{ "complete": true, "report": { "bucket": "vn_S_NARR_reg", "label": "Strong Narrator Style", "definition": { "id": "vn_S_NARR_reg", "category": "talking_style", "head": "vn_S_NARR_reg", "pole": "high", "label": "Strong Narrator Style", "floor": 4.0, "canonical_dimension_name": "Narrator Style" }, "recipe_status": "direct_manual_recipe", "manual_provenance": [ { "id": "rank4/style_s_narr/full-v2-sealed/sealed-selected", "mission": "style_s_narr", "format_status": "CANONICAL", "heldout": true, "qualified": false, "source_cfg": 2.5, "source_model": { "label": "Humaneness 1.5 + real-speech rank 4", "repo": "laion/Humaneness-Voice-Large-1.5-RealSpeech-LoRAs", "revision": "fed33e554f094fcdc89fc0124aed2f9d42018714", "rank": 4 }, "metrics": { "ce": 6.319868102073669, "genuineness": 4.028861799240112, "wer": 0.0009381109481915933, "identity": 0.5398599266260863, "voiceclap": -0.13518258369527758, "fitness": 0.7894878415514077, "steering_v2": 0.9287453572134443, "raw_target": 3.6034338998794557, "target_cdf": 0.9287453572134443, "family_hit_all": null, "nominal_hit_all": null, "n": 100, "valid_fraction": 1, "by_arm": { "reference": { "n": 50, "ce": 6.356364698410034, "genuineness": 3.9897620773315428, "wer": 0.0, "identity": 0.5398599266260863, "fitness": 0.7862508683259827, "steering_v2": 0.9222033286205018 }, "none": { "n": 50, "ce": 6.283371505737304, "genuineness": 4.067961521148682, "wer": 0.0018762218963831866, "identity": null, "fitness": 0.7927248147768328, "steering_v2": 0.9352873858063867 } } } } ], "selected_wording": { "id": "manual-9ea232963c9d", "general": "Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.", "inline": [ "Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.", "Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.", "End with sharp, defined final consonants, giving the narration a clean, authoritative stop." ], "cfg": 2.5, "recipe": "P06", "strategy": "local-Gemma-mutation", "timing_syntax": "canonical", "canonical_labels": true, "neutral_event": "keep", "parent": "cfg-2.5", "origin": "exact_frozen_manual_recipe" }, "selection": { "manual_no_adapter": { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": 1.8402343476164893, "target_raw": 4.048805677890778, "ce": 6.202515125274658, "genuineness": 4.076496934890747, "identity": 0.5526344582438469, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.013450938463211059, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "manual_adapter": { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": 1.905078110334897, "target_raw": 4.141871809959412, "ce": 6.155653357505798, "genuineness": 4.018914270401001, "identity": 0.5395399622619153, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.026545434445142745, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "grid": [ { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": 1.8402343476164893, "target_raw": 4.048805677890778, "ce": 6.202515125274658, "genuineness": 4.076496934890747, "identity": 0.5526344582438469, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.013450938463211059, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.25 }, "n": 20, "target_z": 1.8046875088250176, "target_raw": 3.99778755903244, "ce": 6.1588304281234745, "genuineness": 4.192289876937866, "identity": 0.5516191884875298, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.014466208219528199, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.0 }, "n": 20, "target_z": 1.7886718668238808, "target_raw": 3.9748013257980346, "ce": 6.179878520965576, "genuineness": 4.204842007160186, "identity": 0.5458768345415592, "wer": 0.01423611111111111, "valid_fraction": 0.95, "identity_delta": -0.020208562165498732, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": 1.7585937149594888, "target_raw": 3.93163206577301, "ce": 6.116932821273804, "genuineness": 4.123410391807556, "identity": 0.4998421385884285, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.06624325811862945, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.5 }, "n": 20, "target_z": 1.7312500062294243, "target_raw": 3.8923873782157896, "ce": 6.150093507766724, "genuineness": 4.096894133090973, "identity": 0.5139976121485234, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.05208778455853462, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": 1.6845703142520256, "target_raw": 3.8253909826278685, "ce": 6.099061799049378, "genuineness": 4.011853754520416, "identity": 0.4386576645076275, "wer": 0.01, "valid_fraction": 1, "identity_delta": -0.12742773219943046, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 1.6664062601228142, "target_raw": 3.7993212699890138, "ce": 5.969000601768494, "genuineness": 3.887745034694672, "identity": 0.3832904249429703, "wer": 0.01840686274509804, "valid_fraction": 0.95, "identity_delta": -0.18279497176408768, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 1.6316406492687985, "target_raw": 3.749424397945404, "ce": 6.071262502670288, "genuineness": 3.9886324048042296, "identity": 0.44963691346347334, "wer": 0.012222222222222221, "valid_fraction": 1, "identity_delta": -0.11644848324358463, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": 1.905078110334897, "target_raw": 4.141871809959412, "ce": 6.155653357505798, "genuineness": 4.018914270401001, "identity": 0.5395399622619153, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.026545434445142745, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.25 }, "n": 20, "target_z": 1.8761718709768302, "target_raw": 4.100384521484375, "ce": 6.18464412689209, "genuineness": 4.116663634777069, "identity": 0.5318096876144409, "wer": 0.0061111111111111106, "valid_fraction": 1, "identity_delta": -0.034275709092617034, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.0 }, "n": 20, "target_z": 1.8578124953279318, "target_raw": 4.074034476280213, "ce": 6.168361639976501, "genuineness": 4.086852490901947, "identity": 0.5697233989834786, "wer": 0.003125, "valid_fraction": 1, "identity_delta": 0.0036380022764205934, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.0 }, "n": 20, "target_z": 1.8406249740440659, "target_raw": 4.049366319179535, "ce": 6.208869314193725, "genuineness": 4.148671436309814, "identity": 0.5590439960360527, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.007041400671005249, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.5 }, "n": 20, "target_z": 1.83515625389339, "target_raw": 4.041517412662506, "ce": 6.186629796028138, "genuineness": 4.14490578174591, "identity": 0.5773625567555427, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.011277160048484803, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.0 }, "n": 20, "target_z": 1.828515687683581, "target_raw": 4.031986629962921, "ce": 6.1715387344360355, "genuineness": 4.137374556064605, "identity": 0.5696623027324677, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.0035769060254096986, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.5 }, "n": 20, "target_z": 1.821093768947832, "target_raw": 4.021334421634674, "ce": 6.157241916656494, "genuineness": 4.100973510742188, "identity": 0.5576882809400558, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.008397115767002106, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.0 }, "n": 20, "target_z": 1.8109375150544418, "target_raw": 4.0067577958106995, "ce": 6.231903052330017, "genuineness": 4.216138887405395, "identity": 0.5632766455411911, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.002808751165866852, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.25 }, "n": 20, "target_z": 1.7980468512503203, "target_raw": 3.988256645202637, "ce": 6.191792511940003, "genuineness": 4.079321169853211, "identity": 0.5547879874706269, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.011297409236431123, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.5 }, "n": 20, "target_z": 1.7421874714484724, "target_raw": 3.9080852270126343, "ce": 6.168758821487427, "genuineness": 4.086852359771728, "identity": 0.5499290138483047, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.016156382858753204, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.25 }, "n": 20, "target_z": 1.713085943794314, "target_raw": 3.8663176536560058, "ce": 6.216811943054199, "genuineness": 4.179423999786377, "identity": 0.5641189590096474, "wer": 0.0061111111111111106, "valid_fraction": 1, "identity_delta": -0.0019664376974105834, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": 1.842968740915423, "target_raw": 4.0527301788330075, "ce": 6.134605240821839, "genuineness": 4.010755455493927, "identity": 0.4824974060058594, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.08358799070119857, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.5 }, "n": 20, "target_z": 1.833203146673205, "target_raw": 4.0387142419815065, "ce": 6.156844758987427, "genuineness": 4.002910375595093, "identity": 0.500364363193512, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.065721033513546, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": 1.816796861632694, "target_raw": 4.015167343616485, "ce": 6.134605264663696, "genuineness": 4.072417509555817, "identity": 0.4998321649618447, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.06625323174521328, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 1.7933593839547972, "target_raw": 3.9815290212631225, "ce": 6.108394503593445, "genuineness": 3.971687042713165, "identity": 0.47012153901159764, "wer": 0.023823529411764705, "valid_fraction": 0.9, "identity_delta": -0.09596385769546031, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.0 }, "n": 20, "target_z": 1.7812500062294243, "target_raw": 3.96414920091629, "ce": 6.1322224855422975, "genuineness": 4.0488823533058165, "identity": 0.49943932592868806, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.0666460707783699, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": 1.7644531279849325, "target_raw": 3.940041708946228, "ce": 6.089729142189026, "genuineness": 3.9236753821372985, "identity": 0.4705475814640522, "wer": 0.020426556587547297, "valid_fraction": 1, "identity_delta": -0.09553781524300575, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 3.0 }, "n": 20, "target_z": 1.7593750093441363, "target_raw": 3.9327534079551696, "ce": 6.0841692924499515, "genuineness": 3.9982033133506776, "identity": 0.48935286067426204, "wer": 0.003125, "valid_fraction": 1, "identity_delta": -0.07673253603279591, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 3.0 }, "n": 20, "target_z": 1.751171879282729, "target_raw": 3.9209799766540527, "ce": 6.1242798328399655, "genuineness": 4.003067314624786, "identity": 0.49138830080628393, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.074697095900774, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": 1.6763671924965173, "target_raw": 3.8136175632476808, "ce": 6.014670944213867, "genuineness": 3.9820425271987916, "identity": 0.3962954182177782, "wer": 0.020116959064327485, "valid_fraction": 1, "identity_delta": -0.16978997848927974, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 1.6544921873053304, "target_raw": 3.782221758365631, "ce": 6.050810050964356, "genuineness": 3.934658443927765, "identity": 0.4139593871310353, "wer": 0.0059649122807017545, "valid_fraction": 1, "identity_delta": -0.15212600957602262, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 4.0 }, "n": 20, "target_z": 1.634765602548117, "target_raw": 3.75390944480896, "ce": 6.024599289894104, "genuineness": 3.951917552947998, "identity": 0.4806344099342823, "wer": 0.008204334365325076, "valid_fraction": 1, "identity_delta": -0.08545098677277566, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 1.5871093528076763, "target_raw": 3.6855114579200743, "ce": 6.001565527915955, "genuineness": 3.884293270111084, "identity": 0.44119606986641885, "wer": 0.01768640350877193, "valid_fraction": 1, "identity_delta": -0.12488932684063911, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 1.5824218772062544, "target_raw": 3.678783822059631, "ce": 6.026783537864685, "genuineness": 3.8853915452957155, "identity": 0.39732993617653845, "wer": 0.02408668730650155, "valid_fraction": 0.95, "identity_delta": -0.16875546053051949, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" } ], "development_standard": [ { "scale": 0.0, "n": 20, "target_z": 1.657031267390476, "target_raw": 3.7858659386634828, "ce": 6.208075022697448, "genuineness": 4.300237834453583, "identity": 0.5660853967070579, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.5, "n": 20, "target_z": 1.6474609282207535, "target_raw": 3.772130239009857, "ce": 6.177098560333252, "genuineness": 4.309024214744568, "identity": 0.6025837153196335, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.036498318612575534, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.75, "n": 20, "target_z": 1.5906250072676615, "target_raw": 3.6905572533607485, "ce": 6.1588304281234745, "genuineness": 4.259443533420563, "identity": 0.5698005467653274, "wer": 0.003125, "valid_fraction": 1, "identity_delta": 0.0037151500582695007, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 1.0, "n": 20, "target_z": 1.6394531196790334, "target_raw": 3.760637140274048, "ce": 6.192586779594421, "genuineness": 4.303062009811401, "identity": 0.5841945633292198, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": 0.018109166622161867, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ] }, "heldout": { "standard": { "n": 50, "target_z": 1.5735937465219048, "target_raw": 3.6661133670806887, "ce": 6.230711612701416, "genuineness": 4.349567494392395, "identity": 0.5985677176713944, "wer": 0.012680342159747194, "valid_fraction": 0.98, "identity_delta": 0.0, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "manual": { "n": 50, "target_z": 1.8620312308445206, "target_raw": 4.0800893592834475, "ce": 6.261052646636963, "genuineness": 4.125387306213379, "identity": 0.5303111308813095, "wer": 0.010303030303030302, "valid_fraction": 1, "identity_delta": -0.06825658679008484, "identity_pairs": 50, "gates_pass": false, "gate_scope": "heldout_50_reference_pairs" }, "adapter": { "n": 50, "target_z": 1.886874988579389, "target_raw": 4.115746026039123, "ce": 6.237542333602906, "genuineness": 4.102291498184204, "identity": 0.5742505073547364, "wer": 0.006540560888386975, "valid_fraction": 1, "identity_delta": -0.02431721031665802, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "matched_no_adapter": { "n": 50, "target_z": 1.8620312308445206, "target_raw": 4.0800893592834475, "ce": 6.261052646636963, "genuineness": 4.125387306213379, "identity": 0.5303111308813095, "wer": 0.010303030303030302, "valid_fraction": 1, "identity_delta": -0.06825658679008484, "identity_pairs": 50, "gates_pass": false, "gate_scope": "heldout_50_reference_pairs" } }, "adapter_vs_matched_cfg": { "n": 50, "delta_z": 0.024843757734868396, "ci95_z": [ -0.017812472071414794, 0.06749998754115158 ], "language_deltas": { "en": -0.009374934591045792, "de": 0.014062551392749766, "fr": 0.08281245379843705, "nl": 0.07890623935806702, "es": -0.04218752128386607 }, "languages_improved": 3, "clear_gain": false, "delta_ce": -0.023510313034057616, "delta_genuineness": -0.023095808029174804, "delta_wer": -0.0037624694146433275, "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": 0.024843757734868396, "ci95_z": [ -0.017812472071414794, 0.06749998754115158 ], "language_deltas": { "en": -0.009374934591045792, "de": 0.014062551392749766, "fr": 0.08281245379843705, "nl": 0.07890623935806702, "es": -0.04218752128386607 }, "languages_improved": 3, "clear_gain": false, "delta_ce": -0.023510313034057616, "delta_genuineness": -0.023095808029174804, "delta_wer": -0.0037624694146433275, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "manual_vs_standard": { "n": 50, "delta_z": 0.28843748432261573, "ci95_z": [ 0.18890621610154992, 0.40546872015067564 ], "language_deltas": { "en": 0.31249992939985893, "de": 0.19140623520511746, "fr": 0.14687497196759108, "nl": 0.28359380632437725, "es": 0.507812478716134 }, "languages_improved": 5, "clear_gain": true, "delta_ce": 0.030341033935546876, "delta_genuineness": -0.2241801881790161, "delta_wer": -0.002377311856716891, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "route": { "adapter": false, "scale": 0.0, "cfg": 1.0, "reason": "No new configuration passes the held-out gates; retain standard fallback and inspect audio", "fallback_standard": true }, "reference_only": true, "updated": 1793033867.7978427 }, "rows": [ { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "003a5d4ac16072f5ee1cd47df0caa298013e775fafbb4a8c3bd4ddfeeff1263d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d1ce9e8fc32670a063cfd6e98301d330c3a5634064e1f43e664b44e0a050c217", "generation_seconds_batch": 24.56844624702353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3474459648132324, "target_z": 1.3515624558749118, "ce": 6.117925643920898, "genuineness": 3.860914945602417, "identity": 0.15562430024147034, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/003a5d4ac16072f5ee1cd47df0caa298013e775fafbb4a8c3bd4ddfeeff1263d.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "00df615e2e981f4b159cefa94514d46cd3b9eb2176b6d361d13fd6ec270b0c23", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "20c6aa79d1c87dc749784626fb50caae6b720e468fab5900d73958096980c154", "generation_seconds_batch": 22.945691343047656, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.627765655517578, "target_z": 1.5468750051911868, "ce": 5.927301406860352, "genuineness": 3.6914620399475098, "identity": 0.5484312772750854, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/00df615e2e981f4b159cefa94514d46cd3b9eb2176b6d361d13fd6ec270b0c23.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "0158fdfc94f796ec9ac56668c4da011799db897377a2e3fb617b0ae9047b83cf", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "b789c4531f2a128ab089fc7071e50a5bbb5de3afd826ea4e8ae2da10fd4aac47", "generation_seconds_batch": 23.998685345984995, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 5.8875885009765625, "genuineness": 3.8044304847717285, "identity": 0.5971287488937378, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0158fdfc94f796ec9ac56668c4da011799db897377a2e3fb617b0ae9047b83cf.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "01c5707c022a069455fc64e1e4dbc387e87045a0016b2caab8e3f8d0fc7514ed", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "6a185d1b17ac6733a74c0706561abece825930ca8624f8e76f49f4c4d762f096", "generation_seconds_batch": 22.370293823070824, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 6.3641486167907715, "genuineness": 4.042920112609863, "identity": 0.5536336898803711, "hypothesis": "Tim Semsen. Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.1111111111111111, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/01c5707c022a069455fc64e1e4dbc387e87045a0016b2caab8e3f8d0fc7514ed.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "01cbb4b7fe6dd144b313c9054c3928053c3e083958fdbe786575eb1edeace0a0", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2b13db8592d7dcd587c5f4f97a7b1cf7ac53cc1e12569a5980200d88dc2d0acc", "generation_seconds_batch": 17.956596027011983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8968725204467773, "target_z": 1.7343750259559343, "ce": 6.109982967376709, "genuineness": 4.030367851257324, "identity": 0.5133349895477295, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/01cbb4b7fe6dd144b313c9054c3928053c3e083958fdbe786575eb1edeace0a0.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb58e488d74b6d63793bbe61ac95f6ad183c399014c360dee1741314c5e90dc9", "generation_seconds_batch": 17.49339515902102, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165979385375977, "target_z": 1.9218750467206815, "ce": 6.078212261199951, "genuineness": 4.444586753845215, "identity": 0.8224698305130005, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, lis le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "027d49f808e5cade27c23ec84fde4ff2a50c4a5af8c43d74015c0195a6228cb3", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "02aebe368a407a50d6d83f904f3103719453f5f5991652d320aa10b47033b6b8", "generation_seconds_batch": 23.643980256048962, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7398934364318848, "target_z": 1.6249999584705053, "ce": 6.141753673553467, "genuineness": 3.9801597595214844, "identity": 0.5613865256309509, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/027d49f808e5cade27c23ec84fde4ff2a50c4a5af8c43d74015c0195a6228cb3.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "0284ac2777d5ce9cc3f5a54f70340b786161f5f3d2f249c7f6ba730bb9c8d01e", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "6edaf7361eadde09807965db2249942630f1a9191928eb78854e8991de4ecd68", "generation_seconds_batch": 23.314095688052475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5941271781921387, "target_z": 1.5234374195366038, "ce": 5.80021858215332, "genuineness": 3.7416701316833496, "identity": 0.589401364326477, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0284ac2777d5ce9cc3f5a54f70340b786161f5f3d2f249c7f6ba730bb9c8d01e.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "034db33e99803d39ff66630ec788ef0d53470046c254a878f0938514e8da9719", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "3d5f2b0a29bad8a1228079a9f1eef5542f8626ef77e65a3040b7aedcfc986c2a", "generation_seconds_batch": 25.67211418005172, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.0305562019348145, "genuineness": 4.162164688110352, "identity": 0.43446868658065796, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/034db33e99803d39ff66630ec788ef0d53470046c254a878f0938514e8da9719.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "0399cceb62941f6f6cce75e27df9b046195207903f70797b11b88dde9e8e83f9", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "60eb4987dab23edbecd86dd8bbf23b749d4e2acfb72f33d168b04fd6c9882291", "generation_seconds_batch": 23.161296072998084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7398934364318848, "target_z": 1.6249999584705053, "ce": 6.09409761428833, "genuineness": 3.936227321624756, "identity": 0.5767163038253784, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0399cceb62941f6f6cce75e27df9b046195207903f70797b11b88dde9e8e83f9.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": "04ad28f2237d479429e445573118d1540d66adc212f4d5483ac37b221ac23573", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "404c3f2d8a17d4bf6e56b224cc3744ffbc9e99bea9904dafa0cfee91f2611f67", "generation_seconds_batch": 17.37088181998115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165979385375977, "target_z": 1.9218750467206815, "ce": 6.268836498260498, "genuineness": 3.672633647918701, "identity": 0.404083251953125, "hypothesis": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast, um zu verhindern.", "wer": 0.13636363636363635, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/04ad28f2237d479429e445573118d1540d66adc212f4d5483ac37b221ac23573.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "04bf5cc67e22bf888520610e0ae09e98f9780eb131fa9a3274bf1b5fd794ef85", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "47de0cef7cb69156fba995996eae448bfa900f15e189b6fa201f1674843b4e9e", "generation_seconds_batch": 25.67211418005172, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.681767463684082, "target_z": 2.281250031147121, "ce": 6.1020402908325195, "genuineness": 4.15588903427124, "identity": 0.5417146682739258, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/04bf5cc67e22bf888520610e0ae09e98f9780eb131fa9a3274bf1b5fd794ef85.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "04c95d6954a9654b110adf10935ced1e5ddfabdb4da67c85ac46128b994cd19b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca2b6ed2d8a37ca20d4f518f1d3a43edf822d5f5010ec55e6aadc117ecdc795c", "generation_seconds_batch": 26.517314847093076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.457511901855469, "target_z": 2.1250001245884844, "ce": 5.990842819213867, "genuineness": 4.086852550506592, "identity": 0.6249309182167053, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/04c95d6954a9654b110adf10935ced1e5ddfabdb4da67c85ac46128b994cd19b.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "04c9cbf465c64b5fadae5b587fe22fe1202df381010b24f8c15f5ce836848766", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "0ef7d41e9e07f3bb459f54d401c898f14b3c8d5a75a0cc3eb12967f3506b782b", "generation_seconds_batch": 18.00094519497361, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2332563400268555, "target_z": 1.9687502180298475, "ce": 6.292664527893066, "genuineness": 4.149612903594971, "identity": 0.4121069610118866, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/04c9cbf465c64b5fadae5b587fe22fe1202df381010b24f8c15f5ce836848766.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "054596b845e75d5a2cbc5a78414d9fb5ce7b894bf65c40633ad5b96144909dd3", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "c8ec767d34b84b9ee79bb6bd4518656176b4da25de530c74003497e8fb090c1b", "generation_seconds_batch": 22.254985745996237, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.681767463684082, "target_z": 2.281250031147121, "ce": 6.078212261199951, "genuineness": 4.143336772918701, "identity": 0.6714049577713013, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/054596b845e75d5a2cbc5a78414d9fb5ce7b894bf65c40633ad5b96144909dd3.mp3" }, { "id": "heldout-de-0", "language": "de", "sentences": [ "Deine Nachricht kam, als ich schon weg war.", "Ich las sie zweimal... dann legte ich das Handy weg.", "Sag mir, wann wir reden können." ], "text": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal... dann legte ich das Handy weg. Sag mir, wann wir reden können.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-0", "key": "059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ddd207c2a526d1781da421227a0c7a94e14436a95a2a8527d8e7a87965bd5a2", "generation_seconds_batch": 20.248717847978696, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.212892532348633, "target_z": 1.2578124454925381, "ce": 6.41180419921875, "genuineness": 4.607763767242432, "identity": 0.5220597982406616, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "063923c3a91e39106faac9b02df615554f640ab4fdf7c3d531579f9ab4de3f85", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c0e20ab55360a65fe12bd16526e757ec4beb4720d767049e76196d7aad12ee1b", "generation_seconds_batch": 24.23364193190355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9662113189697266, "target_z": 1.085937481830846, "ce": 6.046441555023193, "genuineness": 4.042920112609863, "identity": 0.06533986330032349, "hypothesis": "Zietmeerbrug. De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/063923c3a91e39106faac9b02df615554f640ab4fdf7c3d531579f9ab4de3f85.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": "0649679aeb6c7e8036897159654729a48f973b4cdffd7de51b37705c9d70305d", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "00342a20f720480aef0f7c3ec7b6200700b941c8d30fd35d0fdab9fe90b14cfd", "generation_seconds_batch": 18.090373842976987, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9192981719970703, "target_z": 1.7500000830589895, "ce": 6.173524379730225, "genuineness": 4.093128204345703, "identity": 0.7507503032684326, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0649679aeb6c7e8036897159654729a48f973b4cdffd7de51b37705c9d70305d.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "0655ca444220e8db1bf8b2c2d1999538b22111506f829cab05f78162e0f5c3fe", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f8cece9cbc4f33da08c99ec47700e042667509d5cb3aa20bb782e642fda478c4", "generation_seconds_batch": 23.52070540492423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7398934364318848, "target_z": 1.6249999584705053, "ce": 6.197352409362793, "genuineness": 3.8797430992126465, "identity": 0.2538608908653259, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0655ca444220e8db1bf8b2c2d1999538b22111506f829cab05f78162e0f5c3fe.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "0717c1f150dfb7f57b2d4afb681b21acd2e8f6ee46c496b97db74ff02dec4d18", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "13a37e526c4313c9f8256c3d084a924b429619490530242dada25687f2b60d3b", "generation_seconds_batch": 22.649762853980064, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.422391414642334, "target_z": 0.7070313421435664, "ce": 5.287916660308838, "genuineness": 4.030367851257324, "identity": 0.3837984800338745, "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_S_NARR_reg/0717c1f150dfb7f57b2d4afb681b21acd2e8f6ee46c496b97db74ff02dec4d18.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "073da5504e7d4a4d56cb5148cb6f9f4d704bd96d3adf6f3640051ddfe99f5155", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "c383ce10e812a7149c790d353316a851d97a244f370bd69766b01d2373c02b20", "generation_seconds_batch": 25.050629122997634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.173524379730225, "genuineness": 4.005263805389404, "identity": 0.32330965995788574, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/073da5504e7d4a4d56cb5148cb6f9f4d704bd96d3adf6f3640051ddfe99f5155.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "07bd1a5705f31ba6cb7c6fd2b910b4a0295a73d3e31f73ec77d979526664c8c8", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "bfc26e9d1b8984cafd81994ddb8a2e82700ae784438ddb310b4c7cd487fd42dc", "generation_seconds_batch": 23.99903186794836, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7511062622070312, "target_z": 1.6328124870220329, "ce": 6.157639026641846, "genuineness": 4.080576419830322, "identity": 0.6883283853530884, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/07bd1a5705f31ba6cb7c6fd2b910b4a0295a73d3e31f73ec77d979526664c8c8.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "07d921c30998d501de27e9bc7f4e945cbdb47f429bd976cbc4d2907f4cb7f27a", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "188eb17ccfa66946749b9c08492b4828b0055439c5d62c1678bd29cdbae6424d", "generation_seconds_batch": 30.98853924300056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.098702430725098, "target_z": 1.8749998754115158, "ce": 6.062326908111572, "genuineness": 4.369274139404297, "identity": 0.7052990794181824, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/07d921c30998d501de27e9bc7f4e945cbdb47f429bd976cbc4d2907f4cb7f27a.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "084d88935d3cb6bd2adf780113f77a699cf385e3537dbe1852d59b426ffdb9d1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2749f6f4d85daf34ebf32584c976993464b05f14da6d5da99d15d30344b6e8b2", "generation_seconds_batch": 23.077090003993362, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.076276779174805, "target_z": 1.8593748183084604, "ce": 6.0305562019348145, "genuineness": 4.143336772918701, "identity": 0.5250720977783203, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/084d88935d3cb6bd2adf780113f77a699cf385e3537dbe1852d59b426ffdb9d1.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "085260d645964879ed88d22510d260159bef93fd5ce80133622623d73097b132", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "f05ca193cabe81d0d53569c84a38b2ed2cac0df53f9abdd3d884943c0ba72134", "generation_seconds_batch": 23.03402772604022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7286806106567383, "target_z": 1.6171874299189777, "ce": 6.324435234069824, "genuineness": 4.168440818786621, "identity": 0.405673086643219, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/085260d645964879ed88d22510d260159bef93fd5ce80133622623d73097b132.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "09290c2a74e4672ef982719593eb4db74bad10ac13501562f3f86fc6f89b0a75", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "b95d1934e70692a8c180246f4ae9585873057b90b4093fea83ebb26cbddf853d", "generation_seconds_batch": 21.6687438840745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.252951145172119, "genuineness": 4.055472373962402, "identity": 0.4362379312515259, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/09290c2a74e4672ef982719593eb4db74bad10ac13501562f3f86fc6f89b0a75.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": "0950534d853060afd32bcc648f26068af00aabef95b6b4901c4b66d26e120b26", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The road turns left behind the bridge.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "63d5e39f8bd3ea2e8ec20c55a152d581f947c37a1ce12f017fb1d4feb93dedb5", "generation_seconds_batch": 20.27465147699695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8071701526641846, "target_z": 1.671874963661692, "ce": 6.340320587158203, "genuineness": 3.7793264389038086, "identity": 0.24169646203517914, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0950534d853060afd32bcc648f26068af00aabef95b6b4901c4b66d26e120b26.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "0a1970507c04f1533f820c00f72655e729924946dd02a56d586ced0e54baaf9e", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "cbab188fd37afd1f0810ef2bc705e82ab2ccf04a23f826427571ad47eb483a59", "generation_seconds_batch": 17.61245301202871, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4707865715026855, "target_z": 1.4374999377057578, "ce": 5.959072113037109, "genuineness": 4.15588903427124, "identity": 0.42870521545410156, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0a1970507c04f1533f820c00f72655e729924946dd02a56d586ced0e54baaf9e.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "0ade6b4836392263bafae13c266488b965e23f5121e6091b40a102e5aa3f6308", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "7acc25950bdb88b35e0cb319f410da7dfaee3cc8ffae408b02d9270ccdf248bb", "generation_seconds_batch": 24.37203737697564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.457511901855469, "target_z": 2.1250001245884844, "ce": 5.863759994506836, "genuineness": 3.7667744159698486, "identity": 0.42969846725463867, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0ade6b4836392263bafae13c266488b965e23f5121e6091b40a102e5aa3f6308.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "0b208d7b2851fc184851a8b33018ac9c1a3bcb01155e790c7d8c196910d686c7", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "12ac3922eae3dc3ce1045b88728ca5e4c7ae24a9b179c3c1c1b19f861370c99f", "generation_seconds_batch": 17.61831695807632, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8968725204467773, "target_z": 1.7343750259559343, "ce": 6.046441555023193, "genuineness": 4.457139015197754, "identity": 0.6825061440467834, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0b208d7b2851fc184851a8b33018ac9c1a3bcb01155e790c7d8c196910d686c7.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "0b277dc71101cecbf4de326fc12a0b7b8e61659423d7347f7c139541eaf0f873", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "91ec8aba46899b0aa81357ee79f533f3ca5fe0a4fff2144308e3a9bf33505ba8", "generation_seconds_batch": 17.12073755392339, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.840808629989624, "target_z": 1.695312549316275, "ce": 6.300607204437256, "genuineness": 3.9927117824554443, "identity": 0.5968074202537537, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0b277dc71101cecbf4de326fc12a0b7b8e61659423d7347f7c139541eaf0f873.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": "0b8385ee55f513dc4204f2c3c0da440aaa8e4070619f857166ccb56c552bc17f", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4840815d7ec87b134285c5fc61881ba203fadff2499a9812b6b24f303b1b0e82", "generation_seconds_batch": 23.234204292064533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.986574649810791, "target_z": 1.7968749221321973, "ce": 6.221180438995361, "genuineness": 4.319066047668457, "identity": 0.5955477952957153, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0b8385ee55f513dc4204f2c3c0da440aaa8e4070619f857166ccb56c552bc17f.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "0b8e8b8885f3e28c30d86fc6e678e278d384ea414fe2bd0a1f8cb7111ffb4f24", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "d373e030291fe06e17a05ab3df141ebbf7154059a3c6ffae61dc250d91f78bcc", "generation_seconds_batch": 24.15808807394933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.457511901855469, "target_z": 2.1250001245884844, "ce": 6.403861999511719, "genuineness": 4.21237325668335, "identity": 0.5889077186584473, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0b8e8b8885f3e28c30d86fc6e678e278d384ea414fe2bd0a1f8cb7111ffb4f24.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "0b953c1fccc16640b84161cdbf56d58400c261c286a161506b7680028df47dc0", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4af0c4763687fb645800711141a3c8f82a62877cd2ba2b44f13a6c633fc63d38", "generation_seconds_batch": 24.042246051016264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.358658790588379, "target_z": 1.3593749844264396, "ce": 5.518253803253174, "genuineness": 3.8169827461242676, "identity": 0.5901753902435303, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0b953c1fccc16640b84161cdbf56d58400c261c286a161506b7680028df47dc0.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "0b9a2743c5242d55a1223ce8c1b12954adb5988147c4495f905192ee20a62ed2", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4949e42fc91beab262c2517ea10b89cf3f34d467e27861e5f13fd9bdff2630cd", "generation_seconds_batch": 24.7399111809209, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.340320587158203, "genuineness": 3.9048471450805664, "identity": 0.6085933446884155, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0b9a2743c5242d55a1223ce8c1b12954adb5988147c4495f905192ee20a62ed2.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "0be503cb03940cb7248eb218993a5dc61ea93a88612d3fcc656989e6acdb44d2", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "1e5c66b412e047681aa8aa1dc17ff28aa636509bcc29ee1b77e266169da212f0", "generation_seconds_batch": 27.649475675076246, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.547214031219482, "target_z": 2.1875000207647473, "ce": 5.824047088623047, "genuineness": 4.061748504638672, "identity": 0.6660298109054565, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0be503cb03940cb7248eb218993a5dc61ea93a88612d3fcc656989e6acdb44d2.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "0c1fa836e83f1b94d5d7ea06d01edcfb55a12cb102759654fa51ec7b9338684b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "b61c763da15b686bb49d477563239c6be4fd1ddb96b188fe4afb11344e645ab5", "generation_seconds_batch": 24.352012494928204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7623190879821777, "target_z": 1.6406250155735604, "ce": 6.356205940246582, "genuineness": 4.281409740447998, "identity": 0.613715648651123, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0c1fa836e83f1b94d5d7ea06d01edcfb55a12cb102759654fa51ec7b9338684b.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "0c234ada1930eac4d135079ad60c9b35d1a25ecaf08330e0009cdda602be69ff", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "33938105a65238a26ca51775d0d2859f0e461546aa57da74b3004eda112456b3", "generation_seconds_batch": 24.77069671510253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.143553733825684, "target_z": 1.9062499896176264, "ce": 5.855817794799805, "genuineness": 3.986435890197754, "identity": 0.6919769644737244, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0c234ada1930eac4d135079ad60c9b35d1a25ecaf08330e0009cdda602be69ff.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "0c25bd1c4e186c5f52428d452421a42267c24acf51007171f101e6efbbe339d2", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "3d382f3e05fcba5326f38c8ea691f363b083186c34eeffba7013febd45e6de7b", "generation_seconds_batch": 24.7911069910042, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.098702430725098, "target_z": 1.8749998754115158, "ce": 6.221180438995361, "genuineness": 4.432034492492676, "identity": 0.7041220664978027, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0c25bd1c4e186c5f52428d452421a42267c24acf51007171f101e6efbbe339d2.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "0c29a9f96e53f43228205e50017e6b69498a1805d395dcfb574eb297b3a98c7d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "f839a243fc74b1b974bbce1ad52e6d8d863e29eaa608b87b90b21c03be387742", "generation_seconds_batch": 23.936171105946414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.098702430725098, "target_z": 1.8749998754115158, "ce": 6.117925643920898, "genuineness": 4.086852550506592, "identity": 0.7568280696868896, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0c29a9f96e53f43228205e50017e6b69498a1805d395dcfb574eb297b3a98c7d.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "0c721f9abf1c5febac1d56071401255b72d570070aee370176c8935d694b3e3c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "e8f5c9d2a7d6beddc6d048124f3531b06ed000a596064f03ae7adc508c8adef7", "generation_seconds_batch": 23.51513467996847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8989346027374268, "target_z": 1.0390624766396592, "ce": 5.982900142669678, "genuineness": 4.030367851257324, "identity": -0.03071218729019165, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0c721f9abf1c5febac1d56071401255b72d570070aee370176c8935d694b3e3c.mp3" }, { "id": "heldout-nl-4", "language": "nl", "sentences": [ "Ik schreef het nummer op de envelop.", "Lees het goed... en controleer de laatste twee cijfers.", "Bel me wanneer je even tijd hebt." ], "text": "Ik schreef het nummer op de envelop. Lees het goed... en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-4", "key": "0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "896826a61aee53d09af07e336fe17d86605362c70c4c6cf327c6ca67dcd3d767", "generation_seconds_batch": 18.40612195606809, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7398934364318848, "target_z": 1.6249999584705053, "ce": 6.252951145172119, "genuineness": 4.356721878051758, "identity": 0.7336956858634949, "hypothesis": "Ik schreef het nummer op de envelop, lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "0d2e062695d7d65c0b9d002579307cb91193f25d8bc304485ddcb0d4b16242f0", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8438eb8a5a38d6d20338e96d33270fa4d825a5d8a1fd91c5edd2a2564f447d86", "generation_seconds_batch": 17.782641093013808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.483288764953613, "genuineness": 3.973883628845215, "identity": 0.34976136684417725, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0d2e062695d7d65c0b9d002579307cb91193f25d8bc304485ddcb0d4b16242f0.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "0ebd5e87dd7191e53d6b0557f9a2a11b9d2e6587f4ed2d366bd9c45b333e2c66", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "ca4f9b0de176319ff0548e3643d9699569f2fdb37db2d9443577860ac56a9bd0", "generation_seconds_batch": 22.722333876998164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.547214031219482, "target_z": 2.1875000207647473, "ce": 6.395919322967529, "genuineness": 4.105680465698242, "identity": 0.6720622777938843, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0ebd5e87dd7191e53d6b0557f9a2a11b9d2e6587f4ed2d366bd9c45b333e2c66.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "0ec023cd0d695259938ea920ce14f30179606a05297334e7ca7a3b5fe666d8e4", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "34397355e1d7c408c54f32a61fd1ab92118d4ed51e249c9070d31a7381d52c87", "generation_seconds_batch": 28.86878123297356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0202131271362305, "target_z": 1.8203125077867803, "ce": 6.2767791748046875, "genuineness": 4.337893962860107, "identity": 0.6556719541549683, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0ec023cd0d695259938ea920ce14f30179606a05297334e7ca7a3b5fe666d8e4.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "0ec9a3f5c9e19b03c3c199e7f0b9d5a8748dfd7fcc7e092e5e3f06a9241fa02c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "25fd6c1e5307480db9e1b5ce1f681a3030f2fe89b6f3c5c3cb749cf051ef1b98", "generation_seconds_batch": 23.603207567008212, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.795957326889038, "target_z": 1.6640624351101645, "ce": 6.165581703186035, "genuineness": 3.747946262359619, "identity": 0.4311617910861969, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0ec9a3f5c9e19b03c3c199e7f0b9d5a8748dfd7fcc7e092e5e3f06a9241fa02c.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "0f5e31548393de306ea3302e798051ba59ed7ea2258c14d832a0a7cd39f0b8ce", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "45a6cad431dc49a98322244b68f7d5a0940a3f305860d8b289d8807c94c3e080", "generation_seconds_batch": 23.150010972982273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4707865715026855, "target_z": 1.4374999377057578, "ce": 6.523001670837402, "genuineness": 4.419482707977295, "identity": 0.5532827377319336, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/0f5e31548393de306ea3302e798051ba59ed7ea2258c14d832a0a7cd39f0b8ce.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "10f9ac514f90b3e1114a517490db7a30a096e0d0a7162d50532e3619ec8acc75", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "70b424dce8005c38d4635af1f18be5952a4383e5108262a64db2ec4f5f481117", "generation_seconds_batch": 17.06443342694547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5604889392852783, "target_z": 1.5, "ce": 6.229123115539551, "genuineness": 4.218649387359619, "identity": 0.2406812608242035, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/10f9ac514f90b3e1114a517490db7a30a096e0d0a7162d50532e3619ec8acc75.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "112368fe7f63bcc6dd63f730b6708edec36c15c542c6f357ef4e941e37140e85", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8e2c821ca8a54982620dca75d64612999ded11d38adc40c72b3f572f4180316f", "generation_seconds_batch": 16.12572433904279, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.065064430236816, "target_z": 1.851562621992891, "ce": 6.181467056274414, "genuineness": 3.9299514293670654, "identity": 0.2759363055229187, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/112368fe7f63bcc6dd63f730b6708edec36c15c542c6f357ef4e941e37140e85.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "115c4e1cb7c001f7afd4aae091d09e4fbf160c683c6b78792b634292e9618764", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "8505ebb544a9b22cd87a14cbc1dba3acd687d76a3b212020f12c1ae2d8e29d3c", "generation_seconds_batch": 23.731006611022167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5829145908355713, "target_z": 1.5156250571030552, "ce": 6.165581703186035, "genuineness": 4.118232727050781, "identity": 0.70353102684021, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/115c4e1cb7c001f7afd4aae091d09e4fbf160c683c6b78792b634292e9618764.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "118dcf01ccaba3fe5e5490f53a655b5e22262d976dc68441580532fbff639dd6", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "4bc1414b1efb1583579989b7a50f4de94b9b00d69fcaa8407e96c2f470afc2e6", "generation_seconds_batch": 23.56437352800276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.031425952911377, "target_z": 1.828125036338308, "ce": 5.95112943649292, "genuineness": 3.8671910762786865, "identity": 0.6340813636779785, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/118dcf01ccaba3fe5e5490f53a655b5e22262d976dc68441580532fbff639dd6.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "11a2e587d9fdaa40f753fb395f72f53154bb2ddb5badd365ebe82adad2ea0f7f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "6a22341710ec8228eed2ecf0123589d572b93c2128b07a49fe4a6bbfd9f7daeb", "generation_seconds_batch": 24.535420494969003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.941723585128784, "target_z": 1.7656249740440657, "ce": 6.284721851348877, "genuineness": 4.20609712600708, "identity": 0.255276620388031, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/11a2e587d9fdaa40f753fb395f72f53154bb2ddb5badd365ebe82adad2ea0f7f.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "11c6154985e2afa9590d58c0fc7cb4646e3882ccfecd930a6dfefb8648b4df88", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4f0a13f4c1436044e630d9a470010a630719c2db90fede7c2d46c7746bb20ee9", "generation_seconds_batch": 23.18086227006279, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.999849796295166, "target_z": 1.109375067485429, "ce": 6.046441555023193, "genuineness": 4.13078498840332, "identity": 0.24199970066547394, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/11c6154985e2afa9590d58c0fc7cb4646e3882ccfecd930a6dfefb8648b4df88.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "125c4a78649a4f0122d25caf2cc2ec1184a2344515098adf544575ed3f5e424b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "7ffa7c306fe568c812511e6e7dcb0cbbea95898baf30cd548a3b6e6680c86014", "generation_seconds_batch": 24.3537193860393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.121128082275391, "target_z": 1.890624932514571, "ce": 5.824047088623047, "genuineness": 4.15588903427124, "identity": 0.4871218502521515, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/125c4a78649a4f0122d25caf2cc2ec1184a2344515098adf544575ed3f5e424b.mp3" }, { "id": "heldout-es-0", "language": "es", "sentences": [ "Tu mensaje llegó después de que me fuera.", "Lo leí dos veces... y guardé el teléfono.", "Dime cuándo podemos hablar." ], "text": "Tu mensaje llegó después de que me fuera. Lo leí dos veces... y guardé el teléfono. Dime cuándo podemos hablar.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-0", "key": "12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d55fa7085808673c8885c2c1bb791716db3f459bac059f00acbbe028a01c3cf0", "generation_seconds_batch": 19.176970800966956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.932572841644287, "target_z": 1.062499896176263, "ce": 6.038498878479004, "genuineness": 4.3881025314331055, "identity": 0.8574420809745789, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "132d0d9b9cdcaefa67433cb3a7202a861335e27f9003ce7250a006e3037ab6e0", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "87c1356ccd04a0d6e3da85909258fe82eca85908a481de10060b6d2d2062c8c1", "generation_seconds_batch": 25.709222984034568, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.143553733825684, "target_z": 1.9062499896176264, "ce": 5.919358730316162, "genuineness": 3.948779582977295, "identity": 0.660332202911377, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce", "wer": 0.10526315789473684, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/132d0d9b9cdcaefa67433cb3a7202a861335e27f9003ce7250a006e3037ab6e0.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "132f13ea9c0128cce3ae6709c6c9d84133aa20302a4e76087117dccc444e6aec", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "e8c357a3d6b568531ab705e0556dfa10480933cbe7e54900c59746230dee9e7c", "generation_seconds_batch": 22.722333876998164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.569639682769775, "target_z": 2.2031250778678024, "ce": 6.332377910614014, "genuineness": 4.18099308013916, "identity": 0.45652276277542114, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/132f13ea9c0128cce3ae6709c6c9d84133aa20302a4e76087117dccc444e6aec.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "133f811f70561600f6fb85a1a0e0cf568adce5257fd8903f4390673b93f1c668", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "895296eba6fa43079226e2fd440b35bd3b98aeda79785fdf6c2ec016cf061d70", "generation_seconds_batch": 24.192906149080954, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.964149236679077, "target_z": 1.781250031147121, "ce": 6.165581703186035, "genuineness": 4.080576419830322, "identity": 0.721348762512207, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/133f811f70561600f6fb85a1a0e0cf568adce5257fd8903f4390673b93f1c668.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "13553959efeb53b09cd7309ed3f73096f09a7178c998e7826acf2080eb5ad3d0", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "98e34e11325186053da98041cea6804921d5a9a0aead798db7fe2d8bae8a59c4", "generation_seconds_batch": 26.767446773010306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.736349105834961, "target_z": 0.9257811448784664, "ce": 5.927301406860352, "genuineness": 3.7667744159698486, "identity": 0.5149966478347778, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/13553959efeb53b09cd7309ed3f73096f09a7178c998e7826acf2080eb5ad3d0.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "1377d7779f843cf1f3c335f379cdbfc37b68fe61970c74c5a895b4667efbd4aa", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "48cddef58a602b47de3105f047a91a0627c991ad81e36c0fd990b7a1c8ad0175", "generation_seconds_batch": 25.46429295593407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8520214557647705, "target_z": 1.7031250778678027, "ce": 6.340320587158203, "genuineness": 4.143336772918701, "identity": 0.45451781153678894, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1377d7779f843cf1f3c335f379cdbfc37b68fe61970c74c5a895b4667efbd4aa.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "140132cb69c32a60e69a6ac14fbde934cc810ed3eb761e30aef80b1494f415ee", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6d2a39851fa93257a8e4137f958ba478f181e2521048230d2ac6430039dc6d9", "generation_seconds_batch": 24.43453179509379, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8968725204467773, "target_z": 1.7343750259559343, "ce": 5.839932441711426, "genuineness": 3.9173994064331055, "identity": 0.3388565182685852, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro de la casa.", "wer": 0.17647058823529413, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/140132cb69c32a60e69a6ac14fbde934cc810ed3eb761e30aef80b1494f415ee.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "142b1c2db073f41dbd924d0aac4b8c77cf423fc38462a1f7f0eed7a541e1da22", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "6ef349857acc6fde3d73107911ecc4cc63c7e38e690465ec72a0b73df5aa1e26", "generation_seconds_batch": 23.643980256048962, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.986574649810791, "target_z": 1.7968749221321973, "ce": 5.824047088623047, "genuineness": 3.8044304847717285, "identity": 0.7537498474121094, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/142b1c2db073f41dbd924d0aac4b8c77cf423fc38462a1f7f0eed7a541e1da22.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "14bdf5cf31a75a45404083379cc23e3737d88b4c83296623fcc8ed6250e482d2", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "5c6af2707742ced2d8ec61a99d65f6c00d2929f1c52792be862ca82bf46ad42b", "generation_seconds_batch": 24.5792173279915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8744468688964844, "target_z": 1.718749968852879, "ce": 6.1894097328186035, "genuineness": 4.168440818786621, "identity": 0.43195489048957825, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/14bdf5cf31a75a45404083379cc23e3737d88b4c83296623fcc8ed6250e482d2.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "14d4e609906c41d5845e74b28e0a602e591c4e5f4cc809a6ead82a2643be7e88", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9c079493901285a3e3b0cef688af6f2f6f6c1fb856697135f439da96cf56894b", "generation_seconds_batch": 24.615514816949144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.24500846862793, "genuineness": 4.15588903427124, "identity": 0.6439327001571655, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/14d4e609906c41d5845e74b28e0a602e591c4e5f4cc809a6ead82a2643be7e88.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "14fc26d2e4aebde64816c1a08946f1129746a3029033cbfe710513261879d466", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "1151fd2222bf8f88b84b9dccb9c1ba4c68f4f7af22091313f0f8fce7c14b5381", "generation_seconds_batch": 23.828927029971965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.109915733337402, "target_z": 1.8828127361990015, "ce": 6.205295085906982, "genuineness": 4.105680465698242, "identity": 0.6038397550582886, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/14fc26d2e4aebde64816c1a08946f1129746a3029033cbfe710513261879d466.mp3" }, { "id": "heldout-de-2", "language": "de", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-2", "key": "15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "09aa55b9688ce14330490e0d45e0dea48bf2a9b957f3b427eea3ef11b05f4c05", "generation_seconds_batch": 18.278840276994742, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.627765655517578, "target_z": 1.5468750051911868, "ce": 6.205295085906982, "genuineness": 4.3881025314331055, "identity": 0.49273747205734253, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "15a96f7b0f01fcf60f776907c16eea0bd0ec72b18fd9d40ebdc637fe3c14da4b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "90bb972590208c7a42cda50f32041df66db3a7dc6a4490e324b8b256ba8bdc93", "generation_seconds_batch": 21.40313338499982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.681767463684082, "target_z": 2.281250031147121, "ce": 6.006728172302246, "genuineness": 3.7667744159698486, "identity": 0.5398797988891602, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/15a96f7b0f01fcf60f776907c16eea0bd0ec72b18fd9d40ebdc637fe3c14da4b.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "15b8e1a89fdf7e2cae14ebd06307bb08ed9463fba8140ad0384d71284fc26a07", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "4f20ff6a11110a4d42d92bef889e450a5f5b11b678f87c428f93df5b5ff33a0b", "generation_seconds_batch": 17.823998603969812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.109915733337402, "target_z": 1.8828127361990015, "ce": 6.4515180587768555, "genuineness": 4.193545341491699, "identity": 0.4604358971118927, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/15b8e1a89fdf7e2cae14ebd06307bb08ed9463fba8140ad0384d71284fc26a07.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "15fc285f2ee62d7615ad552ed3dd3b3323de6fc309939697b81556cb25496c4f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "809d964468c47bd8e2b9d12b89f9c184feb988e1b6c297cf2f17896a9398091d", "generation_seconds_batch": 23.25226094899699, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.042638778686523, "target_z": 1.8359375648898355, "ce": 6.38797664642334, "genuineness": 4.362998008728027, "identity": 0.3458503186702728, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/15fc285f2ee62d7615ad552ed3dd3b3323de6fc309939697b81556cb25496c4f.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "1613680e98b61c7901f3f39c0448a39f165e53b25b546dec89cab30aea0b748a", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "fe6ef23cee55fab4c595e803545a20ff807cf26c3f7d058796452c90098b9d25", "generation_seconds_batch": 26.069628403987736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.292664527893066, "genuineness": 4.287685871124268, "identity": 0.2833709716796875, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1613680e98b61c7901f3f39c0448a39f165e53b25b546dec89cab30aea0b748a.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "16f41622f686157567ec515b62992aa999a2822f6d42f3d3e772cde1a8196b08", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "014b34848746f82c407f579c281b6ffdecc1767d07f0c54332cb385adb2cf10e", "generation_seconds_batch": 22.482404054957442, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8295958042144775, "target_z": 1.6875000207647475, "ce": 6.173524379730225, "genuineness": 4.482243061065674, "identity": 0.6707943677902222, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/16f41622f686157567ec515b62992aa999a2822f6d42f3d3e772cde1a8196b08.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "1707c7534c3042216ed8c8e37cd6787151d993010f860eb6a26c8ba3bc1eff6c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "fbda9c2213101ee15575699ef2e7feedb49f186b7ff86ef9037f4989e316ee70", "generation_seconds_batch": 27.829898354946636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.717468023300171, "target_z": 1.609375067485429, "ce": 6.038498878479004, "genuineness": 4.193545341491699, "identity": 0.5407904386520386, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1707c7534c3042216ed8c8e37cd6787151d993010f860eb6a26c8ba3bc1eff6c.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "17ed712a6718e2dd1a6fda725958626e938acf83b10337275232a8c0ff420fde", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "c2fd3c774f710aad731e1be7bf63a1b6251ae33f3d523e9f5aaadb220bdcf9f9", "generation_seconds_batch": 24.13009187404532, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5156378746032715, "target_z": 1.4687500519118684, "ce": 6.1894097328186035, "genuineness": 4.030367851257324, "identity": 0.31877660751342773, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/17ed712a6718e2dd1a6fda725958626e938acf83b10337275232a8c0ff420fde.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": "18041a10dc3ae783125ebf2e12f9d9a3699c48c6d49b7176bc102da928d5cd67", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "473cdd450ac96d450b4be4582b59dce129066a77a5c664b50f51398f77a31637", "generation_seconds_batch": 20.8017126879422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7847447395324707, "target_z": 1.6562500726766158, "ce": 6.141753673553467, "genuineness": 4.262581825256348, "identity": 0.6628222465515137, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/18041a10dc3ae783125ebf2e12f9d9a3699c48c6d49b7176bc102da928d5cd67.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "184567cebcfdb1279be29f2089e888be65c8a4f9f139df249b0bffb044db66e0", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "001b725d159010000eeaa3f040bc2f60745bda1a9d692412792bce9c4a190ba9", "generation_seconds_batch": 24.39734226593282, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 5.919358730316162, "genuineness": 3.9299514293670654, "identity": 0.42979633808135986, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay detrás.", "wer": 0.058823529411764705, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/184567cebcfdb1279be29f2089e888be65c8a4f9f139df249b0bffb044db66e0.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": "1851b1682cd0f8602ea5222786a555b973e14a8d8000aaa1253fbfcd8267cd9a", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bdb4a944a9c27a8879dcc63a91f51f17747b2d0793fc76aaa3491628ef2dc2be", "generation_seconds_batch": 19.81677030492574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.964149236679077, "target_z": 1.781250031147121, "ce": 6.09409761428833, "genuineness": 4.293961524963379, "identity": 0.7143034338951111, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1851b1682cd0f8602ea5222786a555b973e14a8d8000aaa1253fbfcd8267cd9a.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "18e14435c078b8e93696c5c0ff885146d2df636271d3d4160353c6fc09d3a9a1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dfde0fcbca70b3bec4737fbb61f9e055b16e2dc41d3da610ce28e7898965cd0b", "generation_seconds_batch": 20.45653994102031, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.300532817840576, "target_z": 2.015625057103055, "ce": 6.268836498260498, "genuineness": 4.243753433227539, "identity": 0.47954607009887695, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/18e14435c078b8e93696c5c0ff885146d2df636271d3d4160353c6fc09d3a9a1.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "18e82a4a341db62ee6bf480de20fd35746a358ad40ddfedc4fd2e226cc4680fc", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c8dc87ddb08e9702eddc3fea77a277eace6c79796ae73f4c8c20c3f504ac20fd", "generation_seconds_batch": 23.769994407892227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.932572841644287, "target_z": 1.062499896176263, "ce": 6.0146708488464355, "genuineness": 3.9550557136535645, "identity": 0.20821255445480347, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/18e82a4a341db62ee6bf480de20fd35746a358ad40ddfedc4fd2e226cc4680fc.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "19913087b2ed4e6d61cf7eef366abac423a21e49f665ea5f4aa59ab2ae222609", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "e1ec0f0d1eebad42fe99d8de850290b05f9f7804753e1be947050866bc3e5b8e", "generation_seconds_batch": 23.150010972982273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2241053581237793, "target_z": 1.2656249740440657, "ce": 5.974957466125488, "genuineness": 4.111956596374512, "identity": 0.4355373978614807, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/19913087b2ed4e6d61cf7eef366abac423a21e49f665ea5f4aa59ab2ae222609.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "19f5980f3969d3f6ea3727c3dda4b1869096be632ad2b106aadd6282e52bb403", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "0c0a65293c8d2c75514830a102a937b0cec957a9919108988562cc1973ff3c19", "generation_seconds_batch": 16.766847104998305, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.941723585128784, "target_z": 1.7656249740440657, "ce": 6.213237762451172, "genuineness": 4.375550270080566, "identity": 0.68408203125, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/19f5980f3969d3f6ea3727c3dda4b1869096be632ad2b106aadd6282e52bb403.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "19fd9a671012b7098905173ee95c5c5c7bf9d6d7f419d9349743fa9db9d4d29c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "aa3939286848ea05df3a8d65e13787336b23faab85a973ee9e829db07ba7d826", "generation_seconds_batch": 15.957069286028855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4932122230529785, "target_z": 1.4531249948088132, "ce": 6.1894097328186035, "genuineness": 3.9676074981689453, "identity": 0.3141438364982605, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/19fd9a671012b7098905173ee95c5c5c7bf9d6d7f419d9349743fa9db9d4d29c.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "1a7b3b8334e311dde22971cab242ff8e67993e25aad1f360f1948970fac2e983", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "5724c7e158164d0fcec0a33894c576636171168690415be2fcb6f45ac54401a7", "generation_seconds_batch": 21.535385253024288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5247883796691895, "target_z": 2.171874963661692, "ce": 6.0146708488464355, "genuineness": 3.9299514293670654, "identity": 0.6011221408843994, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1a7b3b8334e311dde22971cab242ff8e67993e25aad1f360f1948970fac2e983.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": "1a8bc54d2b8a381b09e3d6ef42a44b5b44c06b58791b781ac44b0bf90eede0d4", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "43816cde1802adac808887f8f4b9310629bcc1b5eb161ec1935c318187d27a6a", "generation_seconds_batch": 21.08724651299417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.199617385864258, "target_z": 1.9453123001393064, "ce": 6.125868320465088, "genuineness": 4.193545341491699, "identity": 0.6608679294586182, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1a8bc54d2b8a381b09e3d6ef42a44b5b44c06b58791b781ac44b0bf90eede0d4.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "1ace0450a3c8f2d88ee222f12184711388214c913f2213deab770055e288b35e", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d87728d8c64a481b8ad3a07ba568f23ad70026f68c4584be41e5e01866175ff2", "generation_seconds_batch": 16.864902639994398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25568151473999, "target_z": 1.9843749428969448, "ce": 5.935244083404541, "genuineness": 3.898571252822876, "identity": 0.673933744430542, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1ace0450a3c8f2d88ee222f12184711388214c913f2213deab770055e288b35e.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "1b19cf4710153c85893a8183726a95ee75c820d0105ae77686d150dc809b6c9e", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dfe8c787c4c2fba3771d66f7199fc31428294cf8d0f440fe1bddd8f94d956247", "generation_seconds_batch": 23.922300960985012, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6838295459747314, "target_z": 1.585937481830846, "ce": 6.157639026641846, "genuineness": 3.7793264389038086, "identity": 0.23735234141349792, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1b19cf4710153c85893a8183726a95ee75c820d0105ae77686d150dc809b6c9e.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "1b2cc97ceb934064dbc95d70a358c3938fbf797966aa2d72935fbefd2f5d14df", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b95871764bfcdcd883a3bf9ae294b67d824738573491d5d6625280bca18606e4", "generation_seconds_batch": 27.675130528979935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5156378746032715, "target_z": 1.4687500519118684, "ce": 6.229123115539551, "genuineness": 3.973883628845215, "identity": 0.40069058537483215, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1b2cc97ceb934064dbc95d70a358c3938fbf797966aa2d72935fbefd2f5d14df.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "1b33709176e5336c0b8dbae6276cd64da13c623f3768dc2ce00865c269653117", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "bb87767b2950066a8a8a802e1bf68240701ca4716a64cb04f0134d5c4bf6cbef", "generation_seconds_batch": 28.86878123297356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9529364109039307, "target_z": 1.7734375025955935, "ce": 6.229123115539551, "genuineness": 4.168440818786621, "identity": 0.5841593146324158, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1b33709176e5336c0b8dbae6276cd64da13c623f3768dc2ce00865c269653117.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "1b52824ad499cc19e456377d82e886bddd0384f5a46b9541fe2ddd90ec8883ba", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "7a80fd255b80fc42d8b493e68bbee6870f892e769c2620dd6856a6ea6496be89", "generation_seconds_batch": 24.374346121912822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165979385375977, "target_z": 1.9218750467206815, "ce": 6.300607204437256, "genuineness": 3.7542223930358887, "identity": 0.31813672184944153, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1b52824ad499cc19e456377d82e886bddd0384f5a46b9541fe2ddd90ec8883ba.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "1b83508c20d2a704f531c7b70792eed6be63b9c37383d702c1b67a01143e5d31", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c2282865f2319829799aa74f557b32fe79cc7096e541ff361470f56967e331ab", "generation_seconds_batch": 17.62183842493687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2108306884765625, "target_z": 1.9531251609267921, "ce": 5.7128496170043945, "genuineness": 3.8420867919921875, "identity": 0.40348291397094727, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1b83508c20d2a704f531c7b70792eed6be63b9c37383d702c1b67a01143e5d31.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "1baa58f799b776cdfac04123524e290b8b76f3a327ebb12b80250c82b5bae55f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "46cb745faf5e57d05496d5b0c27284a989d472fe8f6add38aec9ba6859ad640c", "generation_seconds_batch": 24.378890975029208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.278107166290283, "target_z": 2.0, "ce": 6.268836498260498, "genuineness": 4.168440818786621, "identity": 0.31613439321517944, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1baa58f799b776cdfac04123524e290b8b76f3a327ebb12b80250c82b5bae55f.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "1c38c7db23bf5f0ee68326aadb90f04f644502769a4e4ac449bf8bfa1c03ba2d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "94114f1041255a2e41470fb3649629bc39ae8f2873b6e80820a790658ec0d2bb", "generation_seconds_batch": 23.018800194957294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.809232234954834, "target_z": 0.9765624143454171, "ce": 5.637393951416016, "genuineness": 4.250029563903809, "identity": 0.7365998029708862, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1c38c7db23bf5f0ee68326aadb90f04f644502769a4e4ac449bf8bfa1c03ba2d.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "1c3b58e9c849049716c15a722176196f4f867ce0197f16f97a0565de0a325414", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "7b731dd038613eab8c752adfae9a0eb3ef3f08d9d175325e9bca026dc1c7902d", "generation_seconds_batch": 18.131777875940315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.042638778686523, "target_z": 1.8359375648898355, "ce": 6.046441555023193, "genuineness": 4.105680465698242, "identity": 0.600714921951294, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1c3b58e9c849049716c15a722176196f4f867ce0197f16f97a0565de0a325414.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "1cb81a3904693a339011b05887f0c5a5716758764cbd9c00cd1b351c99a65788", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "249657cdeeceb78aafc07d0076223ee9a3cdc7d7b4a5a38c14fc356c57b786ef", "generation_seconds_batch": 24.172320017940365, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5247883796691895, "target_z": 2.171874963661692, "ce": 6.292664527893066, "genuineness": 4.030367851257324, "identity": 0.2801186442375183, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1cb81a3904693a339011b05887f0c5a5716758764cbd9c00cd1b351c99a65788.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "1d2059e2f4fe45f2fa989d68cdbe6b476fc2d4062a423546ba19186f77016813", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "de07922ebc5cf50e61805b7831e813836ba44301de6c6a33469b42fe29ddc5bb", "generation_seconds_batch": 23.020264480030164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.403510093688965, "target_z": 1.3906250986325501, "ce": 6.205295085906982, "genuineness": 4.319066047668457, "identity": 0.6373686194419861, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1d2059e2f4fe45f2fa989d68cdbe6b476fc2d4062a423546ba19186f77016813.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "1d3628d508021fce93170d186562b2b79c84174fe32f14ca2513b931fc88ab53", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e4b3150c457651659f7e979bde5464bf7a794453f1195bae6c7b02ab4909659c", "generation_seconds_batch": 23.404025606927462, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4259355068206787, "target_z": 1.4062499896176264, "ce": 6.022613525390625, "genuineness": 3.9048471450805664, "identity": 0.43267473578453064, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1d3628d508021fce93170d186562b2b79c84174fe32f14ca2513b931fc88ab53.mp3" }, { "id": "heldout-es-4", "language": "es", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-4", "key": "1d559e6da25f8bbacc3b53d5880369445633ce7ab0125ca7b07093108b8c8de3", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cef88f5fed49a2c56ac73fc9f98b269858dd722a6bdce6e32c0fd2e6a62afa0b", "generation_seconds_batch": 18.35768487805035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.571701765060425, "target_z": 1.5078125285515276, "ce": 6.332377910614014, "genuineness": 4.394378662109375, "identity": 0.5536035895347595, "hypothesis": "Escribí el número en el sobre. Léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1d559e6da25f8bbacc3b53d5880369445633ce7ab0125ca7b07093108b8c8de3.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "836af0553ae81338679e15046d1f4117a1f3e078c7f8e328620d98bc0f527ded", "generation_seconds_batch": 18.100681693991646, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.818382978439331, "target_z": 1.6796874922132197, "ce": 6.300607204437256, "genuineness": 4.369274139404297, "identity": 0.6148603558540344, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "1df7fc512685268fb11a574c5e4ecec0ea640ea591af72a46d7e748f528902d7", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "d8040b52c77f6023a36f31a5118cf30e5792f6815fdc8d0ede12dba7196a8bc7", "generation_seconds_batch": 16.939567039022222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.59206485748291, "target_z": 2.2187498027348997, "ce": 5.990842819213867, "genuineness": 3.8671910762786865, "identity": 0.5661091804504395, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1df7fc512685268fb11a574c5e4ecec0ea640ea591af72a46d7e748f528902d7.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "1e31cd221fef42235954cb6c604e0841a1644f62f45490316128f2f3b4d7e846", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d7b5e7d575642e5804e255eaa4042ef0eb3574e7a63cae35e61b79c3ae884f12", "generation_seconds_batch": 17.823998603969812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165979385375977, "target_z": 1.9218750467206815, "ce": 6.24500846862793, "genuineness": 4.074300289154053, "identity": 0.14612148702144623, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1e31cd221fef42235954cb6c604e0841a1644f62f45490316128f2f3b4d7e846.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "1e6cb2fd9194a7330ddb0af900f0f85bc0f7ff6f9060a99b899b14008102e471", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 127, "seconds_generated": 10.16, "mp3_sha256": "0b9616f8df763d30322dfd3d0dd82c2171cf511060ca95a1b4b54ec436398cd9", "generation_seconds_batch": 28.31735868891701, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.39023494720459, "target_z": 2.0781249532793185, "ce": 6.300607204437256, "genuineness": 4.162164688110352, "identity": 0.6009567975997925, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1e6cb2fd9194a7330ddb0af900f0f85bc0f7ff6f9060a99b899b14008102e471.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "1ea0f7a563861c44e70428925e8c711475a7db709d2166c61f7bda48fd1b610f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "49066e426ce697e333daa0e6acf7e3f7ab60d6f2cc063b2cbbd9d4c73a0db9d7", "generation_seconds_batch": 19.363819874008186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4483611583709717, "target_z": 1.4218750467206815, "ce": 6.062326908111572, "genuineness": 3.6098732948303223, "identity": 0.3897915184497833, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1ea0f7a563861c44e70428925e8c711475a7db709d2166c61f7bda48fd1b610f.mp3" }, { "id": "heldout-es-4", "language": "es", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-4", "key": "1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e00209c22139324e49da5e539a3f69d72877b9055b034f8323e19653ae53adeb", "generation_seconds_batch": 18.35768487805035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9305107593536377, "target_z": 1.7578124454925381, "ce": 6.41180419921875, "genuineness": 4.20609712600708, "identity": 0.7810122966766357, "hypothesis": "Escribí el número en el. Escribí el número en el sobre, léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.2631578947368421, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "2043fbfb28f8c104fd82fc5f5787542085b88c2ff9195866163497102a4ba731", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "0e33f627b71debb78a905520eaca56ee99db3bae5c679f33f8a991d3489b3ed3", "generation_seconds_batch": 26.84770915308036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.15476655960083, "target_z": 1.914062518169154, "ce": 5.824047088623047, "genuineness": 3.722842216491699, "identity": 0.4338492155075073, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2043fbfb28f8c104fd82fc5f5787542085b88c2ff9195866163497102a4ba731.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "214c4d1a78799113c7f1200d5ba17cf3715291976e3be684255b42e571a90de9", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "1f926c629e625f56ac0dfc1742a38650c6a35720e2bf88e6db3e2e4e205364f5", "generation_seconds_batch": 30.844776742975228, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.717468023300171, "target_z": 1.609375067485429, "ce": 6.372091293334961, "genuineness": 4.256305694580078, "identity": 0.4075937271118164, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/214c4d1a78799113c7f1200d5ba17cf3715291976e3be684255b42e571a90de9.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "2180afd989e3cefa84ec86f618215a1b4b9e61b67f056432a9c3e8fbd47f2b07", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c75af0f6ba43770de81bdb1774bbbddfe383d42ab72b415c955e36e63ed935e9", "generation_seconds_batch": 24.04347878007684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.065064430236816, "target_z": 1.851562621992891, "ce": 6.054384231567383, "genuineness": 4.137060642242432, "identity": 0.49432867765426636, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2180afd989e3cefa84ec86f618215a1b4b9e61b67f056432a9c3e8fbd47f2b07.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "220e6e23882e390dd1f6c179d6d6a5e0fc3dcfd6f18638103fa3f9f09ddd3912", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "3493f2fea39172d6373f69cb0dd57f4bdf7ed2382711399f2cd09c9f3bba8a86", "generation_seconds_batch": 23.915609329007566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165979385375977, "target_z": 1.9218750467206815, "ce": 6.3641486167907715, "genuineness": 3.9927117824554443, "identity": 0.396371066570282, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/220e6e23882e390dd1f6c179d6d6a5e0fc3dcfd6f18638103fa3f9f09ddd3912.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "22538ee9dfb480569304f869cd840b0811148af22134334fb5b9a654f797d66f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "40bad7213c6ec51258fc174145704736ad7fa75464d2324eae43c87ba69ab5ab", "generation_seconds_batch": 28.148975902004167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4707865715026855, "target_z": 1.4374999377057578, "ce": 6.38003396987915, "genuineness": 4.344170093536377, "identity": 0.4373030960559845, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/22538ee9dfb480569304f869cd840b0811148af22134334fb5b9a654f797d66f.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "22797b6b17e2d0476253f2f2bf9ed2e749b76ae941ce25cd93d089aa0c3f2acf", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "4573b60260130c94b79a940da585c0e35ccd250e834cba4758599b5bc71ed3e4", "generation_seconds_batch": 24.895769884926267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1884050369262695, "target_z": 1.937500103823737, "ce": 6.23706579208374, "genuineness": 4.256305694580078, "identity": 0.7248498201370239, "hypothesis": "Et trouver la petite boîte, elle était près, de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.10526315789473684, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/22797b6b17e2d0476253f2f2bf9ed2e749b76ae941ce25cd93d089aa0c3f2acf.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "22b09cb09b870fa41f3c9efcf447d43b07b74e8987f8140a14dff22e8d51ebfd", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "cd534fad363f703b1cf1a340ea5a2b163d8bfb8ac5c503f893425cb5884d2841", "generation_seconds_batch": 24.44029906403739, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.818382978439331, "target_z": 1.6796874922132197, "ce": 6.0146708488464355, "genuineness": 3.848362922668457, "identity": 0.2557585835456848, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/22b09cb09b870fa41f3c9efcf447d43b07b74e8987f8140a14dff22e8d51ebfd.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "230002ff55fea29bbd505838708b92f04f1c31ac65d7efb675137fcc6694b96e", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "271736168f0008b7ac14e20ca3ed81872447d6c64242a9ccada71702e020c4ed", "generation_seconds_batch": 16.979927664040588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.59206485748291, "target_z": 2.2187498027348997, "ce": 6.1894097328186035, "genuineness": 4.20609712600708, "identity": 0.4547595679759979, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/230002ff55fea29bbd505838708b92f04f1c31ac65d7efb675137fcc6694b96e.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "230e8c59796d6cd7013cb5a757da1ed935b989fca9ee20d623fd77c1cfb519cb", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "3f020a2c9986c7b93cb6cbc176ba42c55010bdae266867787b68dac463495a04", "generation_seconds_batch": 24.327331622014754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.278107166290283, "target_z": 2.0, "ce": 6.070269584655762, "genuineness": 4.017816066741943, "identity": 0.7472964525222778, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/230e8c59796d6cd7013cb5a757da1ed935b989fca9ee20d623fd77c1cfb519cb.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "23762226573eb0a6f73e54dff1916e7c6c669604fbd368c7eda9043786274163", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "c70dfeda3688636377d59e10cae59a172cec27f16e08069e68662f5674ade1f8", "generation_seconds_batch": 24.078634591889568, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.345383644104004, "target_z": 2.0468748390732077, "ce": 6.300607204437256, "genuineness": 4.482243061065674, "identity": 0.560089111328125, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/23762226573eb0a6f73e54dff1916e7c6c669604fbd368c7eda9043786274163.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "23e188977f424a3eb859499666a93d6053e9718addfa49b03e64927d6dc4a569", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "a7df6bacb7f69d3394fb11521526ddbc3d0810d75a17d75a2021e66ae6af3c11", "generation_seconds_batch": 23.823509179987013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 5.95112943649292, "genuineness": 3.8546390533447266, "identity": 0.5128351449966431, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/23e188977f424a3eb859499666a93d6053e9718addfa49b03e64927d6dc4a569.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "2408ba67ca44ad254e1cca379229512d275616d27e6cc4483155b9e7c5ff4bc6", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "77ea098f5421ea7bde41a3a56ec98f7ab07044da4dad6d779d76cf75bc097465", "generation_seconds_batch": 17.200668083969504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.717468023300171, "target_z": 1.609375067485429, "ce": 6.0146708488464355, "genuineness": 4.300237655639648, "identity": 0.7520822286605835, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2408ba67ca44ad254e1cca379229512d275616d27e6cc4483155b9e7c5ff4bc6.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "240b4453d62faac19bb5113385fa26cb86a15b30ac0d63396c446359f751ccc7", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "d69364b12c17cb5afd6d39a1b12402716b148c1d4924b54ef880f27af387440a", "generation_seconds_batch": 24.007206290960312, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1884050369262695, "target_z": 1.937500103823737, "ce": 6.1020402908325195, "genuineness": 4.042920112609863, "identity": 0.5305279493331909, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/240b4453d62faac19bb5113385fa26cb86a15b30ac0d63396c446359f751ccc7.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "2492a3005b9c6f72f4f19604877a21ad383d49eb59d241c723c15e6686593edd", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9671fee59cc37f877a34a6b9a463b10ff17dcc651629660295b58a70d0f7c1cb", "generation_seconds_batch": 22.945691343047656, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8372642993927, "target_z": 0.9960937357242362, "ce": 6.054384231567383, "genuineness": 3.8169827461242676, "identity": 0.45550301671028137, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2492a3005b9c6f72f4f19604877a21ad383d49eb59d241c723c15e6686593edd.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "24af46a665719e1c50ae39ade8f837f7100a2404cc0df1f49fab9a559a7046d1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "90a603cd8c977ae587291fdf58d4091c79249e4f37d7df2e4d7a5a58cd382920", "generation_seconds_batch": 24.377227643039078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.863234043121338, "target_z": 1.7109374403013513, "ce": 6.181467056274414, "genuineness": 4.344170093536377, "identity": 0.6059967279434204, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/24af46a665719e1c50ae39ade8f837f7100a2404cc0df1f49fab9a559a7046d1.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "256b6c7a6192bef4817d30bd6c4579b51499caaa78d60d9a245ffb95592f3788", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "73a77ec01a981d3363990106fe35d935c01f9099caf054ba93b678536abce10c", "generation_seconds_batch": 18.09152716305107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4932122230529785, "target_z": 1.4531249948088132, "ce": 6.260893821716309, "genuineness": 3.9801597595214844, "identity": 0.19498595595359802, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/256b6c7a6192bef4817d30bd6c4579b51499caaa78d60d9a245ffb95592f3788.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "256c390a64855e2823dfc6985f6ef930b7bc636e076adf5d8d7e968eee9ec4b3", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "a200edc18c4e96f08abd245455cb654faf67f093f202423dfb1c001506e88b4b", "generation_seconds_batch": 15.755182291031815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7062551975250244, "target_z": 1.6015625389339012, "ce": 6.435632705688477, "genuineness": 4.362998008728027, "identity": 0.434811532497406, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/256c390a64855e2823dfc6985f6ef930b7bc636e076adf5d8d7e968eee9ec4b3.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "25b99f31cfe64e5c83c15aa40df090cf2c22db35c05e9416b158541d01e641e9", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "8dd1c5cfe1056672915a8fdba32c94ed496ebf3ec2b55c55686d715ba3a31713", "generation_seconds_batch": 24.738116881926544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8520214557647705, "target_z": 1.7031250778678027, "ce": 6.24500846862793, "genuineness": 4.143336772918701, "identity": 0.8309350609779358, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/25b99f31cfe64e5c83c15aa40df090cf2c22db35c05e9416b158541d01e641e9.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "25c58632c163bce9a3897c59ca58a058c3b3446c3aa09d84b7a2fbf8253cdc82", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "42689e76a531b73342125ba47125e887f70f5423b4fba2d440a59d95226a92d4", "generation_seconds_batch": 17.53220678807702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.098702430725098, "target_z": 1.8749998754115158, "ce": 6.197352409362793, "genuineness": 3.613011360168457, "identity": 0.19558316469192505, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/25c58632c163bce9a3897c59ca58a058c3b3446c3aa09d84b7a2fbf8253cdc82.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": "25fd7000fff704b96a5780fc1fc948990dc8df88fafa9c38f2f7533aa6fdfd08", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The keys are on the kitchen table.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "936a69e4c8f62ea92c2ced5cf4ec5fd637122de901eb1fb6fe20e828889750df", "generation_seconds_batch": 17.58346256799996, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.098702430725098, "target_z": 1.8749998754115158, "ce": 6.356205940246582, "genuineness": 4.055472373962402, "identity": 0.43313026428222656, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/25fd7000fff704b96a5780fc1fc948990dc8df88fafa9c38f2f7533aa6fdfd08.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "2611ad0e80903458e88e97dcc769766331ccb2112dc5dedcf628bc1bff4f92d1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "e7e99c4f38b661a512e8dab43502da7528da07a25d724edc3b8ca0c90b323e17", "generation_seconds_batch": 25.159852971090004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.372091293334961, "genuineness": 3.798154592514038, "identity": 0.5983890295028687, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2611ad0e80903458e88e97dcc769766331ccb2112dc5dedcf628bc1bff4f92d1.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "269695f10205bc3b0956d34e764c41d776395d3a9b4efaac5738aa96e21ffa39", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "bca0557f17fafe1b18c825f3d73a9a9a05a7054f713ea30b4fae098086b270a5", "generation_seconds_batch": 23.163789084064774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3025949001312256, "target_z": 1.3203125077867803, "ce": 6.24500846862793, "genuineness": 4.595211505889893, "identity": 0.6676745414733887, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/269695f10205bc3b0956d34e764c41d776395d3a9b4efaac5738aa96e21ffa39.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "26c0f03cb81359620f7cb115ec0b123695a849f6c9d587c37a412a4249dd95c9", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7b2eb1c3b040a2eea710ea11f97240df1d554c2802868402f3b07350f934d286", "generation_seconds_batch": 24.45715203799773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 6.173524379730225, "genuineness": 3.9299514293670654, "identity": 0.7563419938087463, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/26c0f03cb81359620f7cb115ec0b123695a849f6c9d587c37a412a4249dd95c9.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "26c3d6622c8978c698fd3844f11a66840c50c53f74ebdcf535b952ea00e415ed", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e48cf2b4b4949a72300e169421cea2aa021dd47e5d00f914f62a9d5cefce7e9d", "generation_seconds_batch": 17.51401129900478, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.336233139038086, "target_z": 1.3437499273233842, "ce": 6.078212261199951, "genuineness": 4.011539936065674, "identity": 0.6659723520278931, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/26c3d6622c8978c698fd3844f11a66840c50c53f74ebdcf535b952ea00e415ed.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c2f1d79a0eb7d02c72a2f2e25ef53030bb4f8797936c5910e2a68b076bc33089", "generation_seconds_batch": 18.23727683199104, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4932122230529785, "target_z": 1.4531249948088132, "ce": 6.4515180587768555, "genuineness": 4.432034492492676, "identity": 0.32049259543418884, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "274c77538eb46fab8d23fac97bcfbd90d2e52d49e206a3224488f64a8ad5f8b2", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d86f86f1dcc034b88f838b50a202564fe8cabf23b075f7baecb306ba16de4f8f", "generation_seconds_batch": 26.517314847093076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 5.863759994506836, "genuineness": 3.9048471450805664, "identity": 0.5637918710708618, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/274c77538eb46fab8d23fac97bcfbd90d2e52d49e206a3224488f64a8ad5f8b2.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": "27b1f2b7b1fdd645a27d1ae16877510c893720dd0f1469514cd1456ea0784b41", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d4300fb3aa9d5461e1f26b032a763a6f93ce69f595adc4c661beb900833fc692", "generation_seconds_batch": 25.765735130989924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2108306884765625, "target_z": 1.9531251609267921, "ce": 6.149696350097656, "genuineness": 4.105680465698242, "identity": 0.6969007253646851, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/27b1f2b7b1fdd645a27d1ae16877510c893720dd0f1469514cd1456ea0784b41.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "28a80e8550212219848b688f96864ebfd901a784b3971bfd5c2c2892cc75b5ad", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "835850c7101cf44dac7c9a9dcc41f83254f38933adbeaf63b338367db59fb9bc", "generation_seconds_batch": 24.357175340061076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4932122230529785, "target_z": 1.4531249948088132, "ce": 6.23706579208374, "genuineness": 4.1998209953308105, "identity": 0.3964059054851532, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/28a80e8550212219848b688f96864ebfd901a784b3971bfd5c2c2892cc75b5ad.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "28c89a87bf174773f2ee98139f948e35712960e6ee0b069a16ee38ca5cfb1cb4", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "443b44d04c47db4dfe36f6ef0e61f0a9497a90c13af97da68caf7d03fd378bfb", "generation_seconds_batch": 17.95437559799757, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.145616054534912, "target_z": 1.2109376064193302, "ce": 6.109982967376709, "genuineness": 3.873466968536377, "identity": 0.7547284960746765, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/28c89a87bf174773f2ee98139f948e35712960e6ee0b069a16ee38ca5cfb1cb4.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": "28fa52f8dc81420f35b7987fcdc2bae6fd7e99507c593b2818d36f24b000bda4", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Someone moved the little blue chair.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4dc7095864c91bbd7f795d3add9ef759570a2f06f8d37803150f5bcb334577a3", "generation_seconds_batch": 17.902906979899853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.053851127624512, "target_z": 1.8437497612054052, "ce": 6.3641486167907715, "genuineness": 4.2374773025512695, "identity": 0.20119544863700867, "hypothesis": "Someone moved the little blue chair. It used to stand here, beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/28fa52f8dc81420f35b7987fcdc2bae6fd7e99507c593b2818d36f24b000bda4.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "29e9b2f3cb34a6597f551bb623225ad1998512e4ddb6d4cc8860f4b692936510", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ad805e29a05b0322c7bc4e799a4b742a1701749ccfc07d0ff65521384dc0f1f1", "generation_seconds_batch": 17.956596027011983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.291382074356079, "target_z": 1.3124999792352525, "ce": 6.300607204437256, "genuineness": 4.369274139404297, "identity": 0.7324367761611938, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/29e9b2f3cb34a6597f551bb623225ad1998512e4ddb6d4cc8860f4b692936510.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "2a40ef6a3ca01d26278e9347a9ea0b61e239d3a0bf4a0514425fa3f5aee039e7", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "146962b43c91ff74ffaf01255284ae77d69b69e0103714841a2658239f7659a4", "generation_seconds_batch": 19.851247643935494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.59206485748291, "target_z": 2.2187498027348997, "ce": 6.332377910614014, "genuineness": 3.9550557136535645, "identity": 0.5968722701072693, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2a40ef6a3ca01d26278e9347a9ea0b61e239d3a0bf4a0514425fa3f5aee039e7.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "2ad55c0e693f6193c42dace2c86a9e6b6b86263f0eeee390c578d3777a3c2715", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "20142625696f70a6845c4d65b416fdf3c294b0151b0e86fefbced7dc6003dab5", "generation_seconds_batch": 23.92989211890381, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5941271781921387, "target_z": 1.5234374195366038, "ce": 6.197352409362793, "genuineness": 4.419482707977295, "identity": 0.7097699046134949, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2ad55c0e693f6193c42dace2c86a9e6b6b86263f0eeee390c578d3777a3c2715.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "2ae68a5d1f17b35c51c5711bbe7ce8551e60b8804d648fdbe0bd2fcc7d2f00ee", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "64b88c2baee4f51af6c63c630ca285e1132e4d3834caa8835c88baaf8afe86fe", "generation_seconds_batch": 22.983637726982124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8744468688964844, "target_z": 1.718749968852879, "ce": 6.070269584655762, "genuineness": 4.068024158477783, "identity": 0.6540046334266663, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2ae68a5d1f17b35c51c5711bbe7ce8551e60b8804d648fdbe0bd2fcc7d2f00ee.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": "2b9694e7bc6245f3167d7ffb064b627555b2f1dc6eb4253a321ece917a2b6f5b", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c5e040ecd3eb5b994a4744a113ba978d286986eea4e8f5f806a914bccd41fafb", "generation_seconds_batch": 18.17393783503212, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.502363204956055, "target_z": 2.1562502387945948, "ce": 6.141753673553467, "genuineness": 3.9927117824554443, "identity": 0.7261509895324707, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2b9694e7bc6245f3167d7ffb064b627555b2f1dc6eb4253a321ece917a2b6f5b.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "2ba4116c4fdb717668fcc233d2497f4149d2347152aab9ef10e2029ddd1772d5", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "166469955aa10d82f88cf0327a7876671225e7aaa131ee18bda11f0f23e1f698", "generation_seconds_batch": 24.068633335991763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9305107593536377, "target_z": 1.7578124454925381, "ce": 5.935244083404541, "genuineness": 4.143336772918701, "identity": 0.49040043354034424, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2ba4116c4fdb717668fcc233d2497f4149d2347152aab9ef10e2029ddd1772d5.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "2bea61f1b78a7d757b87fd76df399accb9d1a189615fca29747fb65d4cf683bf", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "10309246a5a910fd54a28ba4b82d16e59e22e0363dc77d333dd182c0db6a563d", "generation_seconds_batch": 18.892858184990473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6802854537963867, "target_z": 0.8867188343567862, "ce": 5.669164657592773, "genuineness": 3.911123275756836, "identity": 0.2207399159669876, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2bea61f1b78a7d757b87fd76df399accb9d1a189615fca29747fb65d4cf683bf.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "2c2da429591a3720afc258c06e5ba963cd0be318a94638f1fc25442e45a48f20", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a07f7a90d13c4c194f1ade7dab261b8b12754ce04a344a054b0203bc89b471f1", "generation_seconds_batch": 23.870866591925733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.031425952911377, "target_z": 1.828125036338308, "ce": 6.133810997009277, "genuineness": 4.055472373962402, "identity": 0.6737977266311646, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2c2da429591a3720afc258c06e5ba963cd0be318a94638f1fc25442e45a48f20.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "2c9d4e4f58530f21532d88aac7989da3f6166b5fb8c2ade0a0f3f21c9847c8c9", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "7ec937b28f4a213152b9e91db45145da1e2d14a2d4b9d5a366911257187afc6a", "generation_seconds_batch": 18.6391249129083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.795957326889038, "target_z": 1.6640624351101645, "ce": 6.252951145172119, "genuineness": 4.011539936065674, "identity": 0.47539687156677246, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2c9d4e4f58530f21532d88aac7989da3f6166b5fb8c2ade0a0f3f21c9847c8c9.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": "2cca1e764fae46502d4c204428c48217e1953bd5531a3aabda2cf1af3cb6731b", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b99c55643a61347d563625224f00c9d7a4850061ecb47aa41918abd525898e6", "generation_seconds_batch": 22.95962944708299, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8295958042144775, "target_z": 1.6875000207647475, "ce": 6.252951145172119, "genuineness": 3.848362922668457, "identity": 0.7668101191520691, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2cca1e764fae46502d4c204428c48217e1953bd5531a3aabda2cf1af3cb6731b.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "2d486d119823a301f3ac23ba99ce5188386ecbefaa7b4163b08cd82c93a30ff1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4cd066c04e0c57f527e13af3e0f804634b953922895e30d7497860450e0627fc", "generation_seconds_batch": 24.261287345085293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.042638778686523, "target_z": 1.8359375648898355, "ce": 6.078212261199951, "genuineness": 4.042920112609863, "identity": 0.7562420964241028, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2d486d119823a301f3ac23ba99ce5188386ecbefaa7b4163b08cd82c93a30ff1.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "2d75be3fe8a70eba373fc207aad626a2818815edebc80fa212820a303c81d8c9", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "0cc6643b439a761e63b2b4eb81c12bde0f1f9a057b803551f71afad14ed37795", "generation_seconds_batch": 24.45715203799773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.816320896148682, "target_z": 2.375000041529495, "ce": 6.141753673553467, "genuineness": 3.8797430992126465, "identity": 0.8084420561790466, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2d75be3fe8a70eba373fc207aad626a2818815edebc80fa212820a303c81d8c9.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "2da9404a3c8427e19a8d4c652690021c86372418a160328acc873a6c18f14cda", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "5e926558a117ac49dae3d797772d8831ed9ce4001aedd24ae49254a08c0828c8", "generation_seconds_batch": 18.09152716305107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.941723585128784, "target_z": 1.7656249740440657, "ce": 6.229123115539551, "genuineness": 3.9801597595214844, "identity": 0.10733120143413544, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2da9404a3c8427e19a8d4c652690021c86372418a160328acc873a6c18f14cda.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "2dbe38432207a40effaa33346e3a48704627bb5c66587196e75e49cbe7d1df48", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "554a30bfa99a56aab1dbf60e047bb17b38d0860e781629b90b36d19117bd546b", "generation_seconds_batch": 21.853655782993883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2241053581237793, "target_z": 1.2656249740440657, "ce": 6.173524379730225, "genuineness": 3.8546390533447266, "identity": 0.2643415927886963, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2dbe38432207a40effaa33346e3a48704627bb5c66587196e75e49cbe7d1df48.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "2dd465ee336c16b6cab1294b975bbba018f0c15db82509dd2fb97ced106f3034", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "1b6aecc290173babffc89c87362e43017f0c0932b81b97a79713bc41cc6475d8", "generation_seconds_batch": 23.258987743989564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8744468688964844, "target_z": 1.718749968852879, "ce": 6.252951145172119, "genuineness": 4.13078498840332, "identity": 0.3422430157661438, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2dd465ee336c16b6cab1294b975bbba018f0c15db82509dd2fb97ced106f3034.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "2dd7254e841fbcc66d24c3e8337b48ee90d08ad2dc2edda8eca08a91487926f0", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "6eeda7694c04c62450d69cee37a6207a7e3663d0528eeb2bd625e5da9859f3bd", "generation_seconds_batch": 23.852780757006258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25568151473999, "target_z": 1.9843749428969448, "ce": 6.308549880981445, "genuineness": 4.20609712600708, "identity": 0.4729112982749939, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2dd7254e841fbcc66d24c3e8337b48ee90d08ad2dc2edda8eca08a91487926f0.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "2df1ab06e23ff215c27b6dd338cd6b28cdb73d2d09d3ecc5579fb5bdac7a838d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "cf85efe989cb1ff0e852bc6e5f27b2cd8cfb401c2b00124f74a945ed00af7a98", "generation_seconds_batch": 27.688737157033756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.076276779174805, "target_z": 1.8593748183084604, "ce": 6.284721851348877, "genuineness": 4.118232727050781, "identity": 0.34582772850990295, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2df1ab06e23ff215c27b6dd338cd6b28cdb73d2d09d3ecc5579fb5bdac7a838d.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "2f25aeee9d52e8d47ece273745c74ceeae38a43e12eca362d78afc086144dae4", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd01ef5661fc4e7b130c7136a6814a9681c8414fe57230037d653234b3f94498", "generation_seconds_batch": 24.941238360013813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.964149236679077, "target_z": 1.781250031147121, "ce": 6.173524379730225, "genuineness": 4.080576419830322, "identity": 0.7484889030456543, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2f25aeee9d52e8d47ece273745c74ceeae38a43e12eca362d78afc086144dae4.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "2fb3d1cf9693fb7f9281f7c61bd2ff49cedbcd682e73743fbb5de0265de0dfaa", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1841bd00dc5850fcdca8b375321f841f88b7ab1d90141340770d1558a5463170", "generation_seconds_batch": 21.01452677801717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9977874755859375, "target_z": 1.804687450683725, "ce": 6.1020402908325195, "genuineness": 4.256305694580078, "identity": 0.30494946241378784, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/2fb3d1cf9693fb7f9281f7c61bd2ff49cedbcd682e73743fbb5de0265de0dfaa.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "3005615518e0b709a577b2d8d67499c912dbd7d04df4384de057cadd7216a7f8", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0605b5193f45a396da7516cecb2011e136cf4627590915848b62cbefb97de841", "generation_seconds_batch": 24.337888716952875, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165979385375977, "target_z": 1.9218750467206815, "ce": 6.1020402908325195, "genuineness": 4.168440818786621, "identity": 0.5397597551345825, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro de la caja.", "wer": 0.17647058823529413, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3005615518e0b709a577b2d8d67499c912dbd7d04df4384de057cadd7216a7f8.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "31982726b7b1e7249129dd5e799ac398504ac5fe08696e1da2e5a628a7b74e29", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6c2877881d1241259bc9c416412e6e9782a8f7d5911bb789e0c6d95f989569f3", "generation_seconds_batch": 24.552437983918935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.358658790588379, "target_z": 1.3593749844264396, "ce": 6.2767791748046875, "genuineness": 3.986435890197754, "identity": 0.29221922159194946, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/31982726b7b1e7249129dd5e799ac398504ac5fe08696e1da2e5a628a7b74e29.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "3283297bac48bb4987f7b00ebe6287b5d71dfe9da3fe44c0f155e350fe790b08", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "c31abf43ec6d36e3b154845a329256625ebc74da61ee2abb5ba9af25888e4c78", "generation_seconds_batch": 21.96367812389508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5156378746032715, "target_z": 1.4687500519118684, "ce": 5.8875885009765625, "genuineness": 3.8671910762786865, "identity": 0.6321632862091064, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3283297bac48bb4987f7b00ebe6287b5d71dfe9da3fe44c0f155e350fe790b08.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "32a47faf0a61e9ea4e0d77f85c1af4994b76309cb21c47fdfd3e765661baaf54", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dbc13b335187c1290c7c06b8a208c1aa20ea24f03f188086e2239cdd266017bc", "generation_seconds_batch": 24.41286914900411, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8520214557647705, "target_z": 1.7031250778678027, "ce": 6.054384231567383, "genuineness": 4.030367851257324, "identity": 0.20834305882453918, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/32a47faf0a61e9ea4e0d77f85c1af4994b76309cb21c47fdfd3e765661baaf54.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "32c7d604c8fdef529382054debec7093b8f986fd017213b0626f1767fe0a80ea", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f478100a6423c4616a994da299d6a22a3d2d0f8482ae904f84ab6771eae02d7f", "generation_seconds_batch": 15.311973260017112, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.348263263702393, "genuineness": 3.550251007080078, "identity": 0.3596608638763428, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/32c7d604c8fdef529382054debec7093b8f986fd017213b0626f1767fe0a80ea.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "33867dbef757ddf4f84e6923ad6993abf456f0d5aa5223e1f190c416853d810b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "cd010146130ca66635aa1025f1080dbdab6df2b8ac31bcaa30c5b957ea448f52", "generation_seconds_batch": 24.109801337006502, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 5.990842819213867, "genuineness": 3.7793264389038086, "identity": 0.49281448125839233, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/33867dbef757ddf4f84e6923ad6993abf456f0d5aa5223e1f190c416853d810b.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "33cb0e1246cf02023cde7e57907da4ab23b53637cc8c8403dcd0345b48a75a6f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "c02d2a43f8d2b881f2e8b911640a4748801c37fe43357a23def0eea94db87e9c", "generation_seconds_batch": 22.649762853980064, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.986574649810791, "target_z": 1.7968749221321973, "ce": 6.125868320465088, "genuineness": 3.9676074981689453, "identity": 0.6599889993667603, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/33cb0e1246cf02023cde7e57907da4ab23b53637cc8c8403dcd0345b48a75a6f.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "33d921157b90e17849c82801d05b76ed0351ad2bd9af6eadf02e83f2e1bad242", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "17ada50e19b83b7d42ec39354e87583a2c10f932b68da056a44ce61d3fc864dc", "generation_seconds_batch": 24.18446216103621, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.795957326889038, "target_z": 1.6640624351101645, "ce": 5.609594821929932, "genuineness": 3.9927117824554443, "identity": 0.3919788599014282, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/33d921157b90e17849c82801d05b76ed0351ad2bd9af6eadf02e83f2e1bad242.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "33db6c79833ff5a61416af88e6375c3b6e2ded6a5e08001701059b567a4719d1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "18360cea1211e4c519cdd4ee65b2a2b7a39d95b6b29fa1583063a5c60647b536", "generation_seconds_batch": 24.405044105951674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.053851127624512, "target_z": 1.8437497612054052, "ce": 6.09409761428833, "genuineness": 3.9550557136535645, "identity": 0.62252277135849, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/33db6c79833ff5a61416af88e6375c3b6e2ded6a5e08001701059b567a4719d1.mp3" }, { "id": "heldout-nl-2", "language": "nl", "sentences": [ "De weg gaat linksaf na de brug.", "Loop verder... tot aan het kleine houten hek.", "Ik wacht aan de andere kant." ], "text": "De weg gaat linksaf na de brug. Loop verder... tot aan het kleine houten hek. Ik wacht aan de andere kant.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-2", "key": "33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e1946a83764eb65e9aa5a3ddd9cb87aefbcc93d2aa705754f3d7a56bb9675f4d", "generation_seconds_batch": 18.00801996293012, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.526850700378418, "target_z": 1.4765625804633962, "ce": 6.221180438995361, "genuineness": 4.381826400756836, "identity": 0.6868905425071716, "hypothesis": "De weg gaat linksaf na de brug. Loopt verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "341c4a03f8b68a65c761434d6a43fe350aa159f9a7b146604318743f3a566c6d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "84ea888a700e86c096f7c60e518df65be04c703e2991e71b6a7890e1de573f19", "generation_seconds_batch": 24.58658734499477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.313807725906372, "target_z": 1.328125036338308, "ce": 6.062326908111572, "genuineness": 4.105680465698242, "identity": 0.5614604949951172, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/341c4a03f8b68a65c761434d6a43fe350aa159f9a7b146604318743f3a566c6d.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "341f3c6bf28601456c722d8a68def4e998ffedeb5d352c850eb827f12c5a319d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b77093a664cd4a5fe244123f0ab3f9b2727133972bc4ad1b3b9bb986a5d7ccb7", "generation_seconds_batch": 22.76867332297843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.547214031219482, "target_z": 2.1875000207647473, "ce": 5.919358730316162, "genuineness": 3.9676074981689453, "identity": 0.7517948746681213, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/341f3c6bf28601456c722d8a68def4e998ffedeb5d352c850eb827f12c5a319d.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1931eaa9396028b0bbec0b135e7bd8e7d5befb08f967d4a9d6c7c43516e369e8", "generation_seconds_batch": 14.94840498897247, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.941723585128784, "target_z": 1.7656249740440657, "ce": 6.324435234069824, "genuineness": 4.036643981933594, "identity": 0.21758538484573364, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "34f9f7706837454e59309f98e9245d4e79813865ade9c1eeeba1a5a4e6ec4414", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "b5e9ede8f746035462b64f26a3f99bd3d287a4807031898d53516d60d5ff8829", "generation_seconds_batch": 18.334531052038074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.244468688964844, "target_z": 1.976562414345417, "ce": 6.062326908111572, "genuineness": 3.9425034523010254, "identity": 0.6749203205108643, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/34f9f7706837454e59309f98e9245d4e79813865ade9c1eeeba1a5a4e6ec4414.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "35cd0f75571a7d4f8d8502942122a86271190fb1007d581e4ca02bea93051ffd", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "4f6a9854417b0e17460cb35de3e3a3180a393267602d4fe40b8520b79a793102", "generation_seconds_batch": 24.13009187404532, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4707865715026855, "target_z": 1.4374999377057578, "ce": 6.213237762451172, "genuineness": 3.8420867919921875, "identity": 0.3384859561920166, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/35cd0f75571a7d4f8d8502942122a86271190fb1007d581e4ca02bea93051ffd.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "36014f8dacffeb530553bc3b8151a0828b026e24577167085fee9534219d4a30", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "82791c5ecc24670974ceb758b1d51ac52734ab02f188e530374f4d58314b23ad", "generation_seconds_batch": 24.5792173279915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.38003396987915, "genuineness": 4.143336772918701, "identity": 0.30563896894454956, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/36014f8dacffeb530553bc3b8151a0828b026e24577167085fee9534219d4a30.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "3625f91b4acbae62773f675059fdf7653928ba1d6cbf4d4a6a1c2ccdec01b095", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8a9ddb2e72f9f1db286bb59d3b0fe9ecdac5eecd14e905fdc1591dc02644f023", "generation_seconds_batch": 24.39734226593282, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8520214557647705, "target_z": 1.7031250778678027, "ce": 5.831989288330078, "genuineness": 3.773050308227539, "identity": 0.52838134765625, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3625f91b4acbae62773f675059fdf7653928ba1d6cbf4d4a6a1c2ccdec01b095.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "364b66be7c846b9dd7e95e6fd68faef71c2eac8e93edf3a7c4c13ca37fae6102", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "773fcf618d9c5a2b62632ea6441b9fd53c68a4b8b73816eba2d415690b52a3ac", "generation_seconds_batch": 27.695405201986432, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.053851127624512, "target_z": 1.8437497612054052, "ce": 6.109982967376709, "genuineness": 3.8671910762786865, "identity": 0.694433331489563, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/364b66be7c846b9dd7e95e6fd68faef71c2eac8e93edf3a7c4c13ca37fae6102.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "367836f8de12ff7e2b6c1e903f6a67dacc4b892975205bebd72c5af7749a4e69", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "1864f71b1e1ff7a6eecd223d8919d7fbcaf33d4f5fae0719b58f2e09fddd3eb4", "generation_seconds_batch": 17.9515318099875, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.941723585128784, "target_z": 1.7656249740440657, "ce": 6.078212261199951, "genuineness": 4.093128204345703, "identity": 0.6515880227088928, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/367836f8de12ff7e2b6c1e903f6a67dacc4b892975205bebd72c5af7749a4e69.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": "371174768f2b75db0d076134e93903cfa8c162cb418f3c501eb92a601e81df0e", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0f945244c8d15f8effedb765221e80e3497100254c9033c7ce95c0d235b5e3bc", "generation_seconds_batch": 17.37088181998115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.908085346221924, "target_z": 1.7421875545074619, "ce": 6.292664527893066, "genuineness": 3.9299514293670654, "identity": 0.4444451332092285, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/371174768f2b75db0d076134e93903cfa8c162cb418f3c501eb92a601e81df0e.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "37ac86c8e82f3b0aa1e9dc98d3321c34899ed6d6d97df652928d8f41ebb27205", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f4ac88caf3644bdbe4ff47330e7673162a57768d0b9b5dcfb72ef987262ea28b", "generation_seconds_batch": 24.881046845926903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.605340003967285, "target_z": 1.5312499480881316, "ce": 6.181467056274414, "genuineness": 4.356721878051758, "identity": 0.6565625071525574, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/37ac86c8e82f3b0aa1e9dc98d3321c34899ed6d6d97df652928d8f41ebb27205.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "3819c8a31e75e71704a5708d3747a2b6a4d04ebb0e72a2a10c9cab0555e9cccf", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "203d614858c88f73e463bc2d73d42002101a51f4b816521a997a239cc285d380", "generation_seconds_batch": 24.07227678806521, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9977874755859375, "target_z": 1.804687450683725, "ce": 6.117925643920898, "genuineness": 3.8546390533447266, "identity": 0.2678094804286957, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3819c8a31e75e71704a5708d3747a2b6a4d04ebb0e72a2a10c9cab0555e9cccf.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "382ba13f1e13fa5b530739818425531dfbc089ba0c51fddd5e52691adecfdd0c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 127, "seconds_generated": 10.16, "mp3_sha256": "46e4ff6e1c01c0138d7f57c700ed2627c9c5cba17ebbc87fa71726b0d90b3c28", "generation_seconds_batch": 24.7399111809209, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.38003396987915, "genuineness": 4.193545341491699, "identity": 0.5234053134918213, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/382ba13f1e13fa5b530739818425531dfbc089ba0c51fddd5e52691adecfdd0c.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "389127f2fa4fce3dacb9f70f615fbc75ce1ce7d332d9406aaca513c595e5ee3f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c4aa39e466d1ee8082312e38ec20ee252d811ef95096cad84bfaafa49e01dac", "generation_seconds_batch": 26.431296546012163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9192981719970703, "target_z": 1.7500000830589895, "ce": 5.974957466125488, "genuineness": 4.055472373962402, "identity": 0.643925666809082, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/389127f2fa4fce3dacb9f70f615fbc75ce1ce7d332d9406aaca513c595e5ee3f.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "38daad827130bc761f3c2370642c2383d0f1f7cc8f774bc77d445c31e3a7eb76", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "cf479872b37b21b9cf917fcc48ba5df3bc30c98f0218b4586dd747fcba60fb23", "generation_seconds_batch": 24.23364193190355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.109915733337402, "target_z": 1.8828127361990015, "ce": 6.23706579208374, "genuineness": 4.118232727050781, "identity": 0.6034606695175171, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/38daad827130bc761f3c2370642c2383d0f1f7cc8f774bc77d445c31e3a7eb76.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "39773d7442f509bb01ab06f37feaaee5ead0d613fdd61d91b9cbc0d712e727a6", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "f0d15149f5ca6b7464690107800cd82ffb1eb68bf38802ee29f5df714dda00a5", "generation_seconds_batch": 24.225607201922685, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.313807725906372, "target_z": 1.328125036338308, "ce": 6.23706579208374, "genuineness": 3.886019229888916, "identity": 0.4168801009654999, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/39773d7442f509bb01ab06f37feaaee5ead0d613fdd61d91b9cbc0d712e727a6.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "3a044783c2fef49901a3d69bf40ee7b59d81713cdcaec8374dfdc2e8973095b8", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "0531d1f48ac5859d2ba5cf9a56b3cc0d8cb74a5daf96a074246b54d2610d2e93", "generation_seconds_batch": 21.860672302893363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.605340003967285, "target_z": 1.5312499480881316, "ce": 6.078212261199951, "genuineness": 4.011539936065674, "identity": 0.6719244718551636, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3a044783c2fef49901a3d69bf40ee7b59d81713cdcaec8374dfdc2e8973095b8.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": "3a92eab22b6206f5eca56b34dec461caf2db3b85b154503dbb0f20d310b24e7a", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "617c9a550dca99b000d7c7a38ae42d4c7afd0c25ea34e452fe75b98236cb4d3f", "generation_seconds_batch": 18.218874246929772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.627765655517578, "target_z": 1.5468750051911868, "ce": 6.109982967376709, "genuineness": 4.18099308013916, "identity": 0.4743616580963135, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3a92eab22b6206f5eca56b34dec461caf2db3b85b154503dbb0f20d310b24e7a.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "3ad1888ef794a5d02a297a7c184620455ec5c68bae4140e0e0bc7ee6b8725578", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "321d4e59c60c5c438c5be1e13f52f8620d844eb93875d908ce56dfa123ccce0a", "generation_seconds_batch": 18.289270548033528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1884050369262695, "target_z": 1.937500103823737, "ce": 6.229123115539551, "genuineness": 4.419482707977295, "identity": 0.6290773153305054, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3ad1888ef794a5d02a297a7c184620455ec5c68bae4140e0e0bc7ee6b8725578.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "3b0d2acd4cce32ae5d47ad2ac7b1166b6eecabf2946f5e21785c248a12b7e02d", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "e26fe909044840757b8861837bd088a7aeb76b893f90e5612a986ecf2314acad", "generation_seconds_batch": 18.573795247008093, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.086154937744141, "genuineness": 3.8546390533447266, "identity": 0.6865333318710327, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3b0d2acd4cce32ae5d47ad2ac7b1166b6eecabf2946f5e21785c248a12b7e02d.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "3b933d83c7a6a28c5cf7913ee9196cd160f3ba34ca7c0a96c0df11aee5ea9b6b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "fedbdb88753f3942e63ca1e54b0df31840278f90b673f85891185b596ff590ef", "generation_seconds_batch": 22.76867332297843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.502363204956055, "target_z": 2.1562502387945948, "ce": 6.141753673553467, "genuineness": 3.8797430992126465, "identity": 0.5811237096786499, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3b933d83c7a6a28c5cf7913ee9196cd160f3ba34ca7c0a96c0df11aee5ea9b6b.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "3bc9d962c0abc7376b5637e2a55ff6247cd778b023b54255c88d5454c2ab8b61", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "74622387441aa52d4c1af07f4cdaaca60fa23e3163fc36c5d0b4b22b2312f678", "generation_seconds_batch": 17.974390335031785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.266894340515137, "target_z": 1.9921874714484724, "ce": 5.998785495758057, "genuineness": 3.8546390533447266, "identity": 0.5192747712135315, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime, cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3bc9d962c0abc7376b5637e2a55ff6247cd778b023b54255c88d5454c2ab8b61.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "3be06af3e692100db55f3e186456a39469591d7b7eff424482891bc9f8579605", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "9045f381eb92991db254a72a3b583d4b9f4bbbe3c7a332ffad6e63f5b74172d9", "generation_seconds_batch": 24.77069671510253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 6.165581703186035, "genuineness": 4.093128204345703, "identity": 0.5672253370285034, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3be06af3e692100db55f3e186456a39469591d7b7eff424482891bc9f8579605.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "3c01b50dd715c89a7805289c501747b934e98df25fdde725d77164b346f70d18", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "42b725a9c6fe6a71312b5fe51a34e470f523ec751ab079d6605c40ce6abd29de", "generation_seconds_batch": 18.892858184990473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.549276113510132, "target_z": 1.4921874714484724, "ce": 6.038498878479004, "genuineness": 4.042920112609863, "identity": 0.508039116859436, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt, sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3c01b50dd715c89a7805289c501747b934e98df25fdde725d77164b346f70d18.mp3" }, { "id": "heldout-nl-1", "language": "nl", "sentences": [ "De sleutels liggen op de keukentafel.", "Ik keek in de la... en daarna naast het raam.", "Neem ze mee als je weggaat." ], "text": "De sleutels liggen op de keukentafel. Ik keek in de la... en daarna naast het raam. Neem ze mee als je weggaat.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-1", "key": "3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7cb43f19bb6114882a25476c172c3f8e077f764ffa8d67f3de06f3c56153e0de", "generation_seconds_batch": 20.45129947399255, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.098702430725098, "target_z": 1.8749998754115158, "ce": 6.268836498260498, "genuineness": 4.507347106933594, "identity": 0.5916045904159546, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "3c6ed00611d3c97839b879958baf7cf194612ced5186a7a50336d0a98e4fa5fb", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "da916e91b251e5ef715ff3d58f6478a8be909a3ea62f9bb2f713848dc7f9c314", "generation_seconds_batch": 22.482404054957442, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4259355068206787, "target_z": 1.4062499896176264, "ce": 6.268836498260498, "genuineness": 4.344170093536377, "identity": 0.4917791783809662, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3c6ed00611d3c97839b879958baf7cf194612ced5186a7a50336d0a98e4fa5fb.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "3cea93dd1ed4441d7dbbdc3a443728f223755efde119a559b197deddbef65a4c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "6ea1d784500812743c020e2ebb649608f5f53c1593b9bef3abe5130101424b4a", "generation_seconds_batch": 18.041863624937832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.435086250305176, "target_z": 2.109375067485429, "ce": 6.054384231567383, "genuineness": 3.886019229888916, "identity": 0.5578917860984802, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3cea93dd1ed4441d7dbbdc3a443728f223755efde119a559b197deddbef65a4c.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "3d00a502e9cb4d4857bfc21c0e83769e69ac12f1793081357562873f65b3f9f4", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "42cc8acfaced692fe1dfe9feb848069217dfcd0bd9e568f1fc8794f87b677f6b", "generation_seconds_batch": 17.60738989210222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.964149236679077, "target_z": 1.781250031147121, "ce": 6.284721851348877, "genuineness": 4.293961524963379, "identity": 0.44491368532180786, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3d00a502e9cb4d4857bfc21c0e83769e69ac12f1793081357562873f65b3f9f4.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "3d2afb4379e28021601c3220dc9c7d0e1bc632b8349f2e1675a984396fae98aa", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fbf751e69a8dcf0ebe28918a6f3e9249ad40820b6f3493b9f7b25ac17de858b8", "generation_seconds_batch": 24.327331622014754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.284721851348877, "genuineness": 4.20609712600708, "identity": 0.5989366173744202, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3d2afb4379e28021601c3220dc9c7d0e1bc632b8349f2e1675a984396fae98aa.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": "3d2c46c7fa02c19dc5a219c3b00f82986c2c3e1c0bee1c223e0b5ce63f6bd5bd", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "129c6e9b3bedea6a4ce49b47d86805439a4ffb5cea70477f592125f4f9a2b2ef", "generation_seconds_batch": 23.78272386395838, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.59206485748291, "target_z": 2.2187498027348997, "ce": 6.0146708488464355, "genuineness": 4.331617832183838, "identity": 0.7394037246704102, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, lis le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3d2c46c7fa02c19dc5a219c3b00f82986c2c3e1c0bee1c223e0b5ce63f6bd5bd.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "3d750068246a664ad9eacc9b24a2f2ae46ff62a0f548a9a8cdae82465e5e0612", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d3a9c194e7c849abf8e06486cd62f4f046c5cd88dd1bfc996524999976e89b76", "generation_seconds_batch": 24.479270800948143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.39023494720459, "target_z": 2.0781249532793185, "ce": 6.078212261199951, "genuineness": 4.21237325668335, "identity": 0.6904385089874268, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3d750068246a664ad9eacc9b24a2f2ae46ff62a0f548a9a8cdae82465e5e0612.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "3d791187e4a844404d41ff2212ad2e7f3bff3cc264da62e2cb608985f3e51e94", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "64918ff548f724f4a30056a446df3bdbd30655953261413c5aed8f118f3bceb0", "generation_seconds_batch": 28.31735868891701, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5247883796691895, "target_z": 2.171874963661692, "ce": 6.324435234069824, "genuineness": 3.998987913131714, "identity": 0.5895318984985352, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3d791187e4a844404d41ff2212ad2e7f3bff3cc264da62e2cb608985f3e51e94.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "3ddfd7b37d3894abe9d5573528939942ed77ab22ee971518dcf675ba5f7aa264", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "5225030cfb2ebfcecaa6d44a503de809188315869824d718e49f9f4e0e9ff103", "generation_seconds_batch": 22.754279759014025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5156378746032715, "target_z": 1.4687500519118684, "ce": 6.23706579208374, "genuineness": 4.3127899169921875, "identity": 0.43861258029937744, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3ddfd7b37d3894abe9d5573528939942ed77ab22ee971518dcf675ba5f7aa264.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "3de4ec06c72ca2bafbfb7d203fe732350b160c368b3e614f1239ab965fc83c53", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06b603584ae214da2c6a74e9a79a996842f7f0bf4892e9a7d5e6e457caeb2408", "generation_seconds_batch": 23.077090003993362, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8520214557647705, "target_z": 1.7031250778678027, "ce": 5.847874641418457, "genuineness": 3.9927117824554443, "identity": 0.5977462530136108, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3de4ec06c72ca2bafbfb7d203fe732350b160c368b3e614f1239ab965fc83c53.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "3e1de1d1ceec5ac6941cf6c77e4feed318a036e001cf625a2b8d64fb72493353", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "3fdc78f809c742eafce6c2d7ad895ee784ca5bd46536e593ddf53abc2b8abe5c", "generation_seconds_batch": 24.423882469069213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4840617179870605, "target_z": 0.7500000830589895, "ce": 5.919358730316162, "genuineness": 3.7040140628814697, "identity": 0.5750401020050049, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3e1de1d1ceec5ac6941cf6c77e4feed318a036e001cf625a2b8d64fb72493353.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "3e4795627fd9f7c87f20aa800f75901b88f70c4038117ec0b8bfe5ddd3f42b48", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9f55e1e4de7df043ffa1bf70e35a53123568519f35a1fced3adb31c9b8561e16", "generation_seconds_batch": 25.622621134039946, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1884050369262695, "target_z": 1.937500103823737, "ce": 6.213237762451172, "genuineness": 3.292933464050293, "identity": 0.26824888586997986, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3e4795627fd9f7c87f20aa800f75901b88f70c4038117ec0b8bfe5ddd3f42b48.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "3e8a89820641cb08a5efd9a1922339877cc1757f933af83bb06308da314f6394", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "40ae8c774d9f0830316b26fad53cabc3d1136c8d0095c9f812b53a23614e8722", "generation_seconds_batch": 22.193334951996803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5941271781921387, "target_z": 1.5234374195366038, "ce": 6.332377910614014, "genuineness": 3.8671910762786865, "identity": 0.6763348579406738, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3e8a89820641cb08a5efd9a1922339877cc1757f933af83bb06308da314f6394.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "3eb1a17b1066693c420a755d9cb63e9ccf247ba5fdcc0f848379aec20c6a8257", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57600d0c964a213ea936290fa39b8bc6b8a2e5785abbc3b2ed78211b9725fe7d", "generation_seconds_batch": 23.896377769997343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.15476655960083, "target_z": 1.914062518169154, "ce": 6.117925643920898, "genuineness": 4.268857479095459, "identity": 0.6566779017448425, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3eb1a17b1066693c420a755d9cb63e9ccf247ba5fdcc0f848379aec20c6a8257.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "3ed86c0a0662be4e465c97487c0861f3b57da0c069384871dc9a3dc7c6a301b7", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "b8bdde668006176386cac882ec8bdd7f4a1eca1f474b884cda8f74462a2292de", "generation_seconds_batch": 17.60738989210222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.526850700378418, "target_z": 1.4765625804633962, "ce": 6.252951145172119, "genuineness": 4.193545341491699, "identity": 0.5363024473190308, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3ed86c0a0662be4e465c97487c0861f3b57da0c069384871dc9a3dc7c6a301b7.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "3ee6de43329dc6a575a921a7eb606d1d46fcf14887fd833be3c5ee4481ac0885", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "b7fb6703b80064b770c22affb3a354e965de8520371055e043f8b1b7e76847d1", "generation_seconds_batch": 24.364615836995654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5380632877349854, "target_z": 1.4843749428969448, "ce": 6.0146708488464355, "genuineness": 3.8922953605651855, "identity": 0.7343630790710449, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3ee6de43329dc6a575a921a7eb606d1d46fcf14887fd833be3c5ee4481ac0885.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "3f326e7c0edc284c47b73be84a52238105684f70e03e746bce5cbcf465be0576", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "db2d4a1aaca0c5fc19a2a4823ea914411cfc24a055552d9d1d055fbc981ec716", "generation_seconds_batch": 23.585260680993088, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.547214031219482, "target_z": 2.1875000207647473, "ce": 6.125868320465088, "genuineness": 4.13078498840332, "identity": 0.7255576848983765, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3f326e7c0edc284c47b73be84a52238105684f70e03e746bce5cbcf465be0576.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "3f3fcdbea406d25ee980fac4b4d55832317d325e7a0d39fad1f3056d883ca3fb", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d033df9b958d41e15c19897dd0cde645d3190cd815582f8016388e2c7affef03", "generation_seconds_batch": 19.37459076696541, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.672616720199585, "target_z": 1.5781249532793185, "ce": 6.117925643920898, "genuineness": 3.9676074981689453, "identity": 0.5703045725822449, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3f3fcdbea406d25ee980fac4b4d55832317d325e7a0d39fad1f3056d883ca3fb.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "3f54050cef0c13cd9bb3589c1a28feb14fbfe1d862be887858874d09fa00b1ab", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "0df8051cc6f135cbf5c7d08f245750a94164d7566d79b5335bb30f8515506bca", "generation_seconds_batch": 30.128549254965037, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 6.1894097328186035, "genuineness": 3.9676074981689453, "identity": 0.4662618637084961, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3f54050cef0c13cd9bb3589c1a28feb14fbfe1d862be887858874d09fa00b1ab.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "3f5cd7c8bf12ba39499954b217d3ea5d8d46eb106baa9536bc49dfa7a0fe0386", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bcba2236736f89876e1d86c47204867b5c835d8cad1608d0fd9d267119869360", "generation_seconds_batch": 25.159852971090004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.443574905395508, "genuineness": 4.300237655639648, "identity": 0.4002898335456848, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3f5cd7c8bf12ba39499954b217d3ea5d8d46eb106baa9536bc49dfa7a0fe0386.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "3f7229d070d2cc6379ab79161f2155083aec6ebd7193c204e16a3b3b2402b1b9", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "7b8bc2344c17af049a4ecb3f0819e87bc0b72fea319a629040371005a2daca72", "generation_seconds_batch": 16.766847104998305, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9529364109039307, "target_z": 1.7734375025955935, "ce": 6.141753673553467, "genuineness": 4.394378662109375, "identity": 0.7450008988380432, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3f7229d070d2cc6379ab79161f2155083aec6ebd7193c204e16a3b3b2402b1b9.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "3f8f205df2f1249f5047976a10716dc5940b1fa863dea21aef55b28941ccfdd2", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "03dc5414c133ffe43e46be1aafe14a091c7ac9bacef00f767f8b5c0a1a71b2dd", "generation_seconds_batch": 23.836258416064084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165979385375977, "target_z": 1.9218750467206815, "ce": 6.205295085906982, "genuineness": 4.243753433227539, "identity": 0.796053946018219, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3f8f205df2f1249f5047976a10716dc5940b1fa863dea21aef55b28941ccfdd2.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "3fc007ec6c71ba8442a18c376af3f5e323a7135c7f8273abd6aad6d77129bb52", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e84e2ea3ca2ae88d01e2b1b3cedd9b0e1ccfb4e537c98caa8b0c534beac6bfc9", "generation_seconds_batch": 23.543066303012893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.547214031219482, "target_z": 2.1875000207647473, "ce": 6.324435234069824, "genuineness": 4.011539936065674, "identity": 0.3334875702857971, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3fc007ec6c71ba8442a18c376af3f5e323a7135c7f8273abd6aad6d77129bb52.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": "3fcc27b92a8eae22e0dd7edd2166c311e6844cb1096c33be522e22e360cf178f", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I wrote the number on the envelope.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fda67a1d537a84c3d7db6d8ef5ac13b97013b46666bde2abc9ea4556dc9f5232", "generation_seconds_batch": 22.508398242061958, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7847447395324707, "target_z": 1.6562500726766158, "ce": 6.348263263702393, "genuineness": 3.8797430992126465, "identity": 0.22875532507896423, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/3fcc27b92a8eae22e0dd7edd2166c311e6844cb1096c33be522e22e360cf178f.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": "407c95e3bc3bff665ba7c02a0f0cf8118ec25271443ee2544c1f2c2ae9cc546d", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "65ca7c26bc19d2e5bca322359fb5696afa18cd0be7589d83e3daea5b169456ba", "generation_seconds_batch": 15.64319321908988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.614490509033203, "target_z": 2.2343748598379554, "ce": 6.372091293334961, "genuineness": 4.068024158477783, "identity": 0.5123088359832764, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/407c95e3bc3bff665ba7c02a0f0cf8118ec25271443ee2544c1f2c2ae9cc546d.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "40cdba85aaeefd4619c6d33ee265ba38fbc1bfbedaf9cfedf490e496563cbd0a", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "586e534460800b5c7b69ed0e17bed4d6313d26b6cb5cba51649f0bc4f38d74f0", "generation_seconds_batch": 17.06443342694547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5156378746032715, "target_z": 1.4687500519118684, "ce": 5.736677646636963, "genuineness": 4.256305694580078, "identity": 0.44059431552886963, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/40cdba85aaeefd4619c6d33ee265ba38fbc1bfbedaf9cfedf490e496563cbd0a.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "40ffd2e00daf2035a1a3392ad9a3dafa1e77f3bf1ceb5ccc7e0a186a81616038", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "8e4abbcc4251839ad7d2b714f964922558748c1c57c8e7701d55522b3e55570b", "generation_seconds_batch": 23.19453237298876, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.031425952911377, "target_z": 1.828125036338308, "ce": 6.213237762451172, "genuineness": 4.13078498840332, "identity": 0.6000416278839111, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt, sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/40ffd2e00daf2035a1a3392ad9a3dafa1e77f3bf1ceb5ccc7e0a186a81616038.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "41dc4fb30b523ccc919c320b8f831a8a1057df0ff9f8a5381b745e58b0e18cdd", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "59eb925dd3f7590becadf650c873890de2d83a01628e7431fda5330f554ab1aa", "generation_seconds_batch": 23.735356541001238, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.726618766784668, "target_z": 2.3125001453532317, "ce": 6.086154937744141, "genuineness": 4.030367851257324, "identity": 0.6056303977966309, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/41dc4fb30b523ccc919c320b8f831a8a1057df0ff9f8a5381b745e58b0e18cdd.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "42fc51665daa831d97bc9ec6751ab62441c5ab56d7f4f002a97976403d6a2baa", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "de07409933fcbc859318d16b2889b1220bed2521b72a0567efe93e49455060ae", "generation_seconds_batch": 25.42722071299795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9192981719970703, "target_z": 1.7500000830589895, "ce": 6.173524379730225, "genuineness": 4.005263805389404, "identity": 0.7770500183105469, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/42fc51665daa831d97bc9ec6751ab62441c5ab56d7f4f002a97976403d6a2baa.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "433bc512dffd0dff33dc080246109cb6d8d777a3d1c000cc2f39865c7e4acc41", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "a3381d0ceccd6528b199099cf895dd23e54412072293ffacfc161e67881da234", "generation_seconds_batch": 18.078224043012597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7623190879821777, "target_z": 1.6406250155735604, "ce": 6.109982967376709, "genuineness": 4.168440818786621, "identity": 0.7326849699020386, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/433bc512dffd0dff33dc080246109cb6d8d777a3d1c000cc2f39865c7e4acc41.mp3" }, { "id": "heldout-nl-2", "language": "nl", "sentences": [ "De weg gaat linksaf na de brug.", "Loop verder... tot aan het kleine houten hek.", "Ik wacht aan de andere kant." ], "text": "De weg gaat linksaf na de brug. Loop verder... tot aan het kleine houten hek. Ik wacht aan de andere kant.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-2", "key": "434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c95509b886ee4acff82c884b92e2a6e7db93029b249cb95944373b85dffe9fbe", "generation_seconds_batch": 18.00801996293012, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1119775772094727, "target_z": 1.1875000207647475, "ce": 6.165581703186035, "genuineness": 4.344170093536377, "identity": 0.5740076303482056, "hypothesis": "De weg gaat linksaf na de brug. Loop verder, tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cb7f2d60f5b801f32643bbcb6306f7f135bea339297c1fd9a766656dff75737b", "generation_seconds_batch": 20.281209568027407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6165528297424316, "target_z": 1.5390624766396592, "ce": 6.0146708488464355, "genuineness": 4.482243061065674, "identity": 0.7190452814102173, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "43ed9d5e4286a90fbe232ca20e6316e5c6295643d0d7f7af911ebc85d3679f3d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "51990ef256e6bad96a3e2fba391c5d73cf048d7b9b160a0c0521085162e4bfa4", "generation_seconds_batch": 24.04347878007684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.457511901855469, "target_z": 2.1250001245884844, "ce": 6.213237762451172, "genuineness": 4.118232727050781, "identity": 0.7363244295120239, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/43ed9d5e4286a90fbe232ca20e6316e5c6295643d0d7f7af911ebc85d3679f3d.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "443b12b375dc49f4855385d3c59f246dc36b916bbddfbcda1605f768d4ca905b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "b8ea19a706986f8e34c0a6149b4153e5db4f93dedced3090c1499848c7fe49de", "generation_seconds_batch": 23.823509179987013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.502363204956055, "target_z": 2.1562502387945948, "ce": 6.133810997009277, "genuineness": 4.017816066741943, "identity": 0.45788949728012085, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/443b12b375dc49f4855385d3c59f246dc36b916bbddfbcda1605f768d4ca905b.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": "447d81d0787a4bf6d97aa4a6d0150398c5d755ce49aa315875665b04a3cf218a", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f82e3bb1710a68ea0c5b90f810b4645f68dd2460056f9ba0294f370e927ccf46", "generation_seconds_batch": 23.353118296014145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.38797664642334, "genuineness": 4.231201171875, "identity": 0.6010233163833618, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/447d81d0787a4bf6d97aa4a6d0150398c5d755ce49aa315875665b04a3cf218a.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "44b2fbf5ab369b0f21463f5435f8e19adbe7606f44448ee748f54d5cf8597a79", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "cf09f9b6304800d8e148170b22c8c0d18c25e31885fb28fd1bc80d51393fa480", "generation_seconds_batch": 24.405044105951674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9437856674194336, "target_z": 1.0703124247277909, "ce": 5.990842819213867, "genuineness": 3.9299514293670654, "identity": 0.5215548276901245, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/44b2fbf5ab369b0f21463f5435f8e19adbe7606f44448ee748f54d5cf8597a79.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "44ce909ebe0ead56d0d161bc2c9c422d4985e1c2a1fe96a6245eac0a3ba53210", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "29c7876b23cf0f7c806991e2f19a5a5b4eef547b6d56db1f3fe6f11f8517a643", "generation_seconds_batch": 24.37181652802974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.278107166290283, "target_z": 2.0, "ce": 6.062326908111572, "genuineness": 3.7918784618377686, "identity": 0.40041589736938477, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/44ce909ebe0ead56d0d161bc2c9c422d4985e1c2a1fe96a6245eac0a3ba53210.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "45287e91932b61a1544a18cd4514bc61c1c60de6fb83822727e4e7b55c31de24", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "ea286142b461a1a78832dfd34707592487fec80295f9aa9774051603eeedf823", "generation_seconds_batch": 23.45346722495742, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.39023494720459, "target_z": 2.0781249532793185, "ce": 6.062326908111572, "genuineness": 3.986435890197754, "identity": 0.7433735132217407, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/45287e91932b61a1544a18cd4514bc61c1c60de6fb83822727e4e7b55c31de24.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "456d676a2bb61b42ab8091fe02e6e3d0cc4e91296b5883caa521bc6f392a6481", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "afcfecdfd8777f3b98d8659656b184ae8a05c9794858867cc7aa0fe746cad408", "generation_seconds_batch": 19.851247643935494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.659341812133789, "target_z": 2.2656249740440657, "ce": 6.348263263702393, "genuineness": 4.124508857727051, "identity": 0.504004180431366, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/456d676a2bb61b42ab8091fe02e6e3d0cc4e91296b5883caa521bc6f392a6481.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "45cb390d768cd0ce96281af9899d953f1bbca9aafb20a59a7b86d1c5f145dae1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "d6e689efce06b0775ebc0fd55c36c712db528aa52723ee63c4935d590960eb94", "generation_seconds_batch": 25.671361043001525, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.502363204956055, "target_z": 2.1562502387945948, "ce": 6.348263263702393, "genuineness": 4.118232727050781, "identity": 0.3272886276245117, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/45cb390d768cd0ce96281af9899d953f1bbca9aafb20a59a7b86d1c5f145dae1.mp3" }, { "id": "heldout-de-4", "language": "de", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-4", "key": "45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57ef71fdfc444b6e449a49f0f10607bc1a367f8cada86d46112b02e2693c6ffe", "generation_seconds_batch": 18.445992250926793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25568151473999, "target_z": 1.9843749428969448, "ce": 6.268836498260498, "genuineness": 4.18726921081543, "identity": 0.5395500659942627, "hypothesis": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "4655a23811fd95a51669c4d9bf428b0c47e0e79e5036b34eab8b3635c207c3c5", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "9ba10cd390f27da8ace46d4f093381e83be4ba8889180bf071a1b6292ea92339", "generation_seconds_batch": 24.565812389017083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.336233139038086, "target_z": 1.3437499273233842, "ce": 6.213237762451172, "genuineness": 3.6789097785949707, "identity": 0.38826286792755127, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/4655a23811fd95a51669c4d9bf428b0c47e0e79e5036b34eab8b3635c207c3c5.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "466173941b088ee12999c7a4f1ca264d7af556f22c47bf0da1fc9d93b9360777", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "318238b5ad33e053e3a31560597c6f91878f344e2be6f94c41e9c89aa1de4511", "generation_seconds_batch": 18.28934576001484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.483288764953613, "genuineness": 4.337893962860107, "identity": 0.5255712866783142, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/466173941b088ee12999c7a4f1ca264d7af556f22c47bf0da1fc9d93b9360777.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "46fbb6fbe8d83651d82ebfd673ccbee00ca5dea2891b78b006fc03846b7e4b64", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "d0af4f7fcf900c9568090b745da3537cb38e82afec699dfaf025dec23a0caec0", "generation_seconds_batch": 24.120895882952027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5829145908355713, "target_z": 1.5156250571030552, "ce": 6.507116317749023, "genuineness": 4.494794845581055, "identity": 0.5318957567214966, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/46fbb6fbe8d83651d82ebfd673ccbee00ca5dea2891b78b006fc03846b7e4b64.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": "47129ff1a2506a7d7449c422f74814a9d62a108fe89dc5b3f06fa518ad4413e9", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a5e71c92df06c662ece5549e960bae3a22cecb6b8aef25fba50d96bac6c6982", "generation_seconds_batch": 25.765735130989924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8295958042144775, "target_z": 1.6875000207647475, "ce": 6.213237762451172, "genuineness": 3.9676074981689453, "identity": 0.7394179105758667, "hypothesis": "De weg gaat linksaf, na de brug. Loop verder, tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/47129ff1a2506a7d7449c422f74814a9d62a108fe89dc5b3f06fa518ad4413e9.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "474971a9e7d445891f81bfb8d993bf84e100a7286c2c0216ebbb9e3d9694e00d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "a90ba26db532bfbe3b2466ca9542336f7d40b388aab4a31a16236235228f24e7", "generation_seconds_batch": 22.904354044003412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2577438354492188, "target_z": 1.2890625596986487, "ce": 6.205295085906982, "genuineness": 4.011539936065674, "identity": 0.3326953053474426, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/474971a9e7d445891f81bfb8d993bf84e100a7286c2c0216ebbb9e3d9694e00d.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "476439597df781b7f4b54518afd2164c99b544c11c38e5d19c4df09a8e3dda83", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "3b869071d6c97390353d550315acb81b006d7acc5fe76d6d4f758686896087ee", "generation_seconds_batch": 23.878003800986335, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 5.8875885009765625, "genuineness": 4.068024158477783, "identity": 0.44266077876091003, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/476439597df781b7f4b54518afd2164c99b544c11c38e5d19c4df09a8e3dda83.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "47734f55289293ddbdb76fc3449e2b2b63117bfd248dd81e08e573ccfea62215", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "f3a38984fceba36e71ef8515bc93020fb02f0b30a0a38091a78c9a2976c577e4", "generation_seconds_batch": 24.190641853027046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.681767463684082, "target_z": 2.281250031147121, "ce": 6.372091293334961, "genuineness": 4.099404335021973, "identity": 0.531740665435791, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/47734f55289293ddbdb76fc3449e2b2b63117bfd248dd81e08e573ccfea62215.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": "479998966c13a7ff4ede64db107d0a268ee7e6692b968fb9425c84c798889804", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "411ee9adbd2a3a1fcc78210b1d13abe4598177c5eb9c780a8375775c2434c7bb", "generation_seconds_batch": 18.16453671292402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.266894340515137, "target_z": 1.9921874714484724, "ce": 6.292664527893066, "genuineness": 4.068024158477783, "identity": 0.47867923974990845, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi, quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/479998966c13a7ff4ede64db107d0a268ee7e6692b968fb9425c84c798889804.mp3" }, { "id": "heldout-fr-3", "language": "fr", "sentences": [ "Quelqu’un a déplacé la petite chaise bleue.", "Elle était ici... près de la grande bibliothèque.", "Nous la remettrons après le déjeuner." ], "text": "Quelqu’un a déplacé la petite chaise bleue. Elle était ici... près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-3", "key": "47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "94b7410fdc471c5150b64184f76cbace66823c6b4557ba1844ef5e19f1e5dcf5", "generation_seconds_batch": 17.881119925063103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.143553733825684, "target_z": 1.9062499896176264, "ce": 6.149696350097656, "genuineness": 4.482243061065674, "identity": 0.7080388069152832, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "4803fb1dc869c259fc55eac21392d349fb65c0444621e09f3863ba5788149213", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e4caef125c72fb45a58ea541561300c1c396c38efd001df70d0e931a860ed9db", "generation_seconds_batch": 23.543066303012893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.2767791748046875, "genuineness": 4.005263805389404, "identity": 0.3615649938583374, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/4803fb1dc869c259fc55eac21392d349fb65c0444621e09f3863ba5788149213.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": "481a26d7ea3652405030b41b492b63c339766c62691df6893a0f92ef25525ecc", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5c31ea93b81acba6567a96f34ca4f4558f0cfc78496200453a1fa8246deee753", "generation_seconds_batch": 18.76157329895068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7398934364318848, "target_z": 1.6249999584705053, "ce": 6.229123115539551, "genuineness": 4.174716949462891, "identity": 0.5767631530761719, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine Houtenhek. Ik wacht aan de andere kant.", "wer": 0.14285714285714285, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/481a26d7ea3652405030b41b492b63c339766c62691df6893a0f92ef25525ecc.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "484c3c923c025cbea057bf9ce143a5b7ce5b1429a5b594c9ae5c1a7bef0e9665", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "de991c23771055317201b216e1f9449f892a153eb94748bd27a107fdba239511", "generation_seconds_batch": 24.378890975029208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165979385375977, "target_z": 1.9218750467206815, "ce": 6.2767791748046875, "genuineness": 4.093128204345703, "identity": 0.36331528425216675, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/484c3c923c025cbea057bf9ce143a5b7ce5b1429a5b594c9ae5c1a7bef0e9665.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "485c48ec2171b68cbbddc89388af5ccb0d6f21add75a280b5d0b6adbbee22f7f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "46156b03bebb0a4320d3390d364cb71aefe5338775cdbe4b84f326620edc765a", "generation_seconds_batch": 23.933880429016426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7286806106567383, "target_z": 1.6171874299189777, "ce": 6.078212261199951, "genuineness": 3.936227321624756, "identity": 0.5998328924179077, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/485c48ec2171b68cbbddc89388af5ccb0d6f21add75a280b5d0b6adbbee22f7f.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "48eb2c651b0559a813cd56e8056824150d9583327a7935573550134844cf503e", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "2a84a450206c6449cea4e7a4b5e93e17199ea23553f755646bf5eadf778242b9", "generation_seconds_batch": 16.939567039022222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.726618766784668, "target_z": 2.3125001453532317, "ce": 5.935244083404541, "genuineness": 3.7793264389038086, "identity": 0.5627875924110413, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/48eb2c651b0559a813cd56e8056824150d9583327a7935573550134844cf503e.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "4911f76f98a091bb307a0e08333a0090f772e6d6c947f773d2ac36157aca4605", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "225ba8cdc2b6bf567f1e71e67967b56e69ec30a9c1ddffa3e5fecc26698e6049", "generation_seconds_batch": 23.30425623198971, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.042638778686523, "target_z": 1.8359375648898355, "ce": 6.308549880981445, "genuineness": 4.256305694580078, "identity": 0.263076514005661, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/4911f76f98a091bb307a0e08333a0090f772e6d6c947f773d2ac36157aca4605.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "49b6e96e589b7ef10c17f73671d05043e889ccb7f04dd3edbba6964b71cfc21b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "613f76acd9d76045829547c59da00f5d56d07f1f5798aa61a857ee7d45755e21", "generation_seconds_batch": 28.148975902004167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8071701526641846, "target_z": 1.671874963661692, "ce": 6.125868320465088, "genuineness": 3.9550557136535645, "identity": 0.3740364611148834, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/49b6e96e589b7ef10c17f73671d05043e889ccb7f04dd3edbba6964b71cfc21b.mp3" }, { "id": "heldout-es-0", "language": "es", "sentences": [ "Tu mensaje llegó después de que me fuera.", "Lo leí dos veces... y guardé el teléfono.", "Dime cuándo podemos hablar." ], "text": "Tu mensaje llegó después de que me fuera. Lo leí dos veces... y guardé el teléfono. Dime cuándo podemos hablar.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-0", "key": "49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6169d22a2d2b12b106fd6c7dd7e4998614b533277beae3013d9e2a0eca9136e3", "generation_seconds_batch": 19.176970800966956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8071701526641846, "target_z": 1.671874963661692, "ce": 6.165581703186035, "genuineness": 4.231201171875, "identity": 0.6558609008789062, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "4a0fecac1d78cfff12c5354f993e7aaf64ac2445ecb0f7bc4b3aaf56d157bb0b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "d48f080f6bad5ef7ed405caf4f67bcfb7da0674b02a0f23ac4b4a9c26ad35090", "generation_seconds_batch": 22.217977733002044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.59206485748291, "target_z": 2.2187498027348997, "ce": 6.173524379730225, "genuineness": 4.055472373962402, "identity": 0.5572501420974731, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/4a0fecac1d78cfff12c5354f993e7aaf64ac2445ecb0f7bc4b3aaf56d157bb0b.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "4a2a9ef85e189871fc5ee82f57c1248bb8ca9854913121e781b9e392b23f730e", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "fe8b6b5284065fd6e091a573a7c22b9d226ca6e2f747bbf9fa80beadf2567fc8", "generation_seconds_batch": 25.709222984034568, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.046441555023193, "genuineness": 4.293961524963379, "identity": 0.6512935161590576, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/4a2a9ef85e189871fc5ee82f57c1248bb8ca9854913121e781b9e392b23f730e.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "4a38ff38f204aaa617d1d1b29fc99f002bf463a6c7a95bebddd8f7fd893afa72", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "015520e55c3a65de0ffc56f901e889086e23c2d67dc35d0648e471868fd966a8", "generation_seconds_batch": 18.131777875940315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.121128082275391, "target_z": 1.890624932514571, "ce": 6.165581703186035, "genuineness": 4.262581825256348, "identity": 0.6976264715194702, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/4a38ff38f204aaa617d1d1b29fc99f002bf463a6c7a95bebddd8f7fd893afa72.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "4ae3e2874aa81c5050ceabf53468fde47d71986ddc3d1a6a4aacda4c39e64205", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e4ad11fbd41ae60aef7b821026490a5ef164db8b1f4f031a7bf3785b734422c6", "generation_seconds_batch": 18.00094519497361, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.300532817840576, "target_z": 2.015625057103055, "ce": 6.316492557525635, "genuineness": 4.137060642242432, "identity": 0.36417317390441895, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/4ae3e2874aa81c5050ceabf53468fde47d71986ddc3d1a6a4aacda4c39e64205.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "4af6cf5260cb27163de3e4e590a34a9c5cb0e7948a44263c7cd78c1cf1c989c2", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "77dd7f39077391ac7c08b8ecaf021cc81d6f5ad973ad5c672b6d5e68529ab419", "generation_seconds_batch": 16.722289137076586, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.605340003967285, "target_z": 1.5312499480881316, "ce": 6.181467056274414, "genuineness": 3.986435890197754, "identity": 0.39287257194519043, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/4af6cf5260cb27163de3e4e590a34a9c5cb0e7948a44263c7cd78c1cf1c989c2.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "4b8675d16f56a6945275376d901d352a30c2180d296e74229b112840ff2b6095", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "67e5dbda0ea67f3875f1bf31a90d2696f0f1c32189531f7ffd0fa6126e79a43e", "generation_seconds_batch": 25.509166829986498, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.502363204956055, "target_z": 2.1562502387945948, "ce": 6.1894097328186035, "genuineness": 4.193545341491699, "identity": 0.6640567183494568, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/4b8675d16f56a6945275376d901d352a30c2180d296e74229b112840ff2b6095.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "4bcac69b936aad1d6cb3efc9a8d1173a09132bb1936dfd73037150d2d69c5ef5", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c9b84d3fcb974fc81d58f6d022700e4d650bd098c505ee8b5256b032e05bd2c", "generation_seconds_batch": 24.308317846036516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.736349105834961, "target_z": 0.9257811448784664, "ce": 5.911416053771973, "genuineness": 3.6412534713745117, "identity": 0.38941070437431335, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/4bcac69b936aad1d6cb3efc9a8d1173a09132bb1936dfd73037150d2d69c5ef5.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "4c035a8860dd444da6eb54bcaf5a987437dd3a9b74254024d12d9f8158fc31d8", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "c284cc15e6bdd9960f3c42bea040576703456076789b10089cbbaa43688e1983", "generation_seconds_batch": 24.479775142041035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.042638778686523, "target_z": 1.8359375648898355, "ce": 6.221180438995361, "genuineness": 4.243753433227539, "identity": 0.7483747005462646, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen eerst samen inkijken.", "wer": 0.16666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/4c035a8860dd444da6eb54bcaf5a987437dd3a9b74254024d12d9f8158fc31d8.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "4ce4065338f544c416e67a56960849abe5b7b13414f14450e62302dcb4c543b1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b9db42545275c641a111e2fcfa8abe61f015a34939191a9edac40466d40a6a0", "generation_seconds_batch": 23.29165158199612, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.284721851348877, "genuineness": 3.6475296020507812, "identity": 0.429347962141037, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/4ce4065338f544c416e67a56960849abe5b7b13414f14450e62302dcb4c543b1.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "4ce9ac0f76ad3d333d73e617f5c96abece594ae4c10c55ae3516d0d82598a37f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c98a20f02a2a64b5cf3a015747312193a0a0d17258c326a7b2b1d74db523e122", "generation_seconds_batch": 23.550526696955785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.121128082275391, "target_z": 1.890624932514571, "ce": 6.221180438995361, "genuineness": 4.20609712600708, "identity": 0.8057488203048706, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/4ce9ac0f76ad3d333d73e617f5c96abece594ae4c10c55ae3516d0d82598a37f.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "4e69f0ad398ad3e9ca669308ff6556d1a17b4c4c29244510ab7d53864d1f3b9a", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "f1c87c8466696c3d421351b8940e1e448d2fafb89a3781c4528a4667c364ecbf", "generation_seconds_batch": 25.022201505023986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.502363204956055, "target_z": 2.1562502387945948, "ce": 6.125868320465088, "genuineness": 3.823258876800537, "identity": 0.6080836057662964, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/4e69f0ad398ad3e9ca669308ff6556d1a17b4c4c29244510ab7d53864d1f3b9a.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": "4ebe060924e3ecc81838aa52c62ee3a5bb7fbcf62eca95a2541e73caffcbaa0a", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f1550b932886f55511cee990f180dfe3859d9f79140f91e554b4caf201d2ec5", "generation_seconds_batch": 23.57263160601724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.300532817840576, "target_z": 2.015625057103055, "ce": 6.3641486167907715, "genuineness": 4.2374773025512695, "identity": 0.5122666358947754, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/4ebe060924e3ecc81838aa52c62ee3a5bb7fbcf62eca95a2541e73caffcbaa0a.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "4ef18eb4a7d67b6df868180bc0fc2480daf4a72323cee2972790327457f84fea", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f69b2e15304ad8d2adcdc4cf70efebd065eb95b546ef0313b1e289013ed963ca", "generation_seconds_batch": 23.2142091980204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2241053581237793, "target_z": 1.2656249740440657, "ce": 5.911416053771973, "genuineness": 3.6851859092712402, "identity": 0.08023367077112198, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/4ef18eb4a7d67b6df868180bc0fc2480daf4a72323cee2972790327457f84fea.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": "4f44c3bb6a0c172cde66263f28c8e3af743605e0e73560e33caa61bb55785155", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "efa54441e0d972cbd8d3841eb4c1a110d7a03cfb7c0c0735e7d69353c6ad99f5", "generation_seconds_batch": 17.951629588962533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.157639026641846, "genuineness": 4.369274139404297, "identity": 0.645390510559082, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/4f44c3bb6a0c172cde66263f28c8e3af743605e0e73560e33caa61bb55785155.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "50680a79f8b9e3e3ad1cc18d8b89bf1d91918394aba7e3c4e63dc299e56a2297", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "70194bb97cb282e1b964cc643c0e250e5f1dc706cca20a98ea1353e015778bd3", "generation_seconds_batch": 25.050629122997634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.749044418334961, "target_z": 2.328125202456287, "ce": 6.213237762451172, "genuineness": 4.061748504638672, "identity": 0.7318305969238281, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/50680a79f8b9e3e3ad1cc18d8b89bf1d91918394aba7e3c4e63dc299e56a2297.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "50a0dba185e2650c76f1cb0ed7a2cbb07fef76eb9c674f6e635e1fa2a0ad6a72", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1e4a1fa2e325aa2c2e2df97bdbbd7942909fe7b6750a9028ebf63e862f05631", "generation_seconds_batch": 24.907093473011628, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.39023494720459, "target_z": 2.0781249532793185, "ce": 5.967014789581299, "genuineness": 3.9425034523010254, "identity": 0.3723175525665283, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/50a0dba185e2650c76f1cb0ed7a2cbb07fef76eb9c674f6e635e1fa2a0ad6a72.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "50c2bb218d5742efd5be67814620887049e18924bfe42a6df8c54055dda6ba27", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "386bc807db6417257d34feb7d9079d19b5576759e1005af975944c9bc737d5c8", "generation_seconds_batch": 26.765301414066926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.941723585128784, "target_z": 1.7656249740440657, "ce": 6.125868320465088, "genuineness": 4.143336772918701, "identity": 0.4827660918235779, "hypothesis": "La portée est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/50c2bb218d5742efd5be67814620887049e18924bfe42a6df8c54055dda6ba27.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "50c8eb40f5f47a8e13e7db20b6bf69e4e5553a0d8ab8a630838598fea87bcd4d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "173612351e0d97393b5a1a17b04d252ef1c1292e123709192fa4df9e5a49b2be", "generation_seconds_batch": 24.377227643039078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.908085346221924, "target_z": 1.7421875545074619, "ce": 6.181467056274414, "genuineness": 3.9550557136535645, "identity": 0.6407783031463623, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/50c8eb40f5f47a8e13e7db20b6bf69e4e5553a0d8ab8a630838598fea87bcd4d.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "50c9ea186cb8f6afcf0b821f028ce8f9027471466a604783f3da4d5221e95cfe", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "33a717da6ab2e2234e32980dd0e2010f096844ec700b61ee35a4872f4689c267", "generation_seconds_batch": 23.66782324796077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.2767791748046875, "genuineness": 3.562803030014038, "identity": 0.3494032621383667, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/50c9ea186cb8f6afcf0b821f028ce8f9027471466a604783f3da4d5221e95cfe.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "50e956e42cbb73426ade6fcb727fd3a5ed7f3149c86c8359c4ce657344f06326", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b70481b980f535410783eeedac827c4c67e4d37f002355312fc7604f0e7ba87", "generation_seconds_batch": 24.258738000062294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8968725204467773, "target_z": 1.7343750259559343, "ce": 6.205295085906982, "genuineness": 4.331617832183838, "identity": 0.6581869721412659, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/50e956e42cbb73426ade6fcb727fd3a5ed7f3149c86c8359c4ce657344f06326.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "512ff947f0caf24ed1b6e7006f827aa7b39578c0438961e093b735ee0d016cc3", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "49972942dde5d06d58b5ca55895be004c963ec1ffb87d90563a32d77baf00d7e", "generation_seconds_batch": 22.254985745996237, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5247883796691895, "target_z": 2.171874963661692, "ce": 6.109982967376709, "genuineness": 4.143336772918701, "identity": 0.6520966291427612, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/512ff947f0caf24ed1b6e7006f827aa7b39578c0438961e093b735ee0d016cc3.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "513c9063acf1ccb02d54dd1f5d67a2fd4d4128695a731b6433f2a1fb64cce672", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "e9399e4f87c44095cfb6af584e5df148ccc1949e24754de7774eaa240fd2d8a9", "generation_seconds_batch": 24.37203737697564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.614490509033203, "target_z": 2.2343748598379554, "ce": 6.229123115539551, "genuineness": 3.8044304847717285, "identity": 0.6553808450698853, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/513c9063acf1ccb02d54dd1f5d67a2fd4d4128695a731b6433f2a1fb64cce672.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "516a61c7b3d4fa5194417db22f9594d8da3c96ea317268e04e5bdee4f2acf586", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "be1ebc23da47e65c5d774584fd0d9e0436848e5846c05df36aa0f23a24df4e01", "generation_seconds_batch": 22.915816162945703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0202131271362305, "target_z": 1.8203125077867803, "ce": 5.911416053771973, "genuineness": 4.369274139404297, "identity": 0.6888467073440552, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/516a61c7b3d4fa5194417db22f9594d8da3c96ea317268e04e5bdee4f2acf586.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "51851deb6cf76827cd51733ad6e804fbaa47cc3694597aa87f8d1ca805f51af1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "dbd97ba4e073248ce3a7cd4697ea68f643667dc3af61e9ee577d3b0763a8ef66", "generation_seconds_batch": 18.05116326804273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.59206485748291, "target_z": 2.2187498027348997, "ce": 6.006728172302246, "genuineness": 4.049196243286133, "identity": 0.36907681822776794, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/51851deb6cf76827cd51733ad6e804fbaa47cc3694597aa87f8d1ca805f51af1.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "52967b93fdabac8eb4166dcbbb50f595002d73c8bfa07db6e2771bc255e32d9a", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "a7743c8cb7a34a7df7f31c42177bafbc5df844731eaa61d2bf926cd024d1a4f6", "generation_seconds_batch": 24.374346121912822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.435086250305176, "target_z": 2.109375067485429, "ce": 6.221180438995361, "genuineness": 3.8797430992126465, "identity": 0.3347730040550232, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/52967b93fdabac8eb4166dcbbb50f595002d73c8bfa07db6e2771bc255e32d9a.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": "52e3a8aabdcdba63b02cbd5ebbab748094ba137cbdb50cdacb8ba2e7a58dddcf", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Escribí el número en el sobre.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d140bc504408f38d0f653c94da04aca3373539231cfaad29a8fb0088e7740020", "generation_seconds_batch": 17.197586350957863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.278107166290283, "target_z": 2.0, "ce": 6.300607204437256, "genuineness": 4.256305694580078, "identity": 0.3820100724697113, "hypothesis": "Escribí el número en el sobre, léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/52e3a8aabdcdba63b02cbd5ebbab748094ba137cbdb50cdacb8ba2e7a58dddcf.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "5322088faf28e08be476654517914acc210070ee6fcfc9cbce406cea7243dac8", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "074fca1e3c00210619aee426e1c7aade4ba85103b5f2eaf98bdf6278336e85d2", "generation_seconds_batch": 24.479270800948143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.681767463684082, "target_z": 2.281250031147121, "ce": 6.062326908111572, "genuineness": 4.281409740447998, "identity": 0.514410138130188, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5322088faf28e08be476654517914acc210070ee6fcfc9cbce406cea7243dac8.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": "53f8002df5615f25e26fe6e963ca186325f94f6801b04ac5528177122dc633db", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The keys are on the kitchen table.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b434400d3e53f61e4e7fa2fdbc73b3b8e814def39f9b9600b7d00153ac884496", "generation_seconds_batch": 24.034143017954193, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.345383644104004, "target_z": 2.0468748390732077, "ce": 6.507116317749023, "genuineness": 4.381826400756836, "identity": 0.32009950280189514, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/53f8002df5615f25e26fe6e963ca186325f94f6801b04ac5528177122dc633db.mp3" }, { "id": "heldout-de-4", "language": "de", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-4", "key": "546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2a8bb554f393ea940cea9a1b0f8eecbc5621ecea512c42a84b7ef240fcd07557", "generation_seconds_batch": 18.445992250926793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.143553733825684, "target_z": 1.9062499896176264, "ce": 6.221180438995361, "genuineness": 3.8671910762786865, "identity": 0.4388725459575653, "hypothesis": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "548e01b7eb469058fe9818b33f72f4e838f9c3beadaf21589e701fb4187b8e51", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "153036efbba98378a08e2c5621f63e6f0bd4944cfc531f94d0faf0c35e80e1cd", "generation_seconds_batch": 24.023838283959776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.749044418334961, "target_z": 2.328125202456287, "ce": 6.23706579208374, "genuineness": 3.9425034523010254, "identity": 0.814665675163269, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/548e01b7eb469058fe9818b33f72f4e838f9c3beadaf21589e701fb4187b8e51.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "54c2c34e04f96bb5fd1586621b8aa61ae145a44942dfc6a8a618e4b74cbdef55", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "4cf3ebf05749f1af89d625a84d3e6465bcfca239059c15230b299bdfede84e91", "generation_seconds_batch": 23.79455260199029, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.795957326889038, "target_z": 1.6640624351101645, "ce": 6.229123115539551, "genuineness": 4.137060642242432, "identity": 0.4902880787849426, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/54c2c34e04f96bb5fd1586621b8aa61ae145a44942dfc6a8a618e4b74cbdef55.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "552cac04dd072abc56293fd14a957d8442ba02625d4f45d4ccf4afb26afb38f6", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad94046f14b26340d0716adc7b24a4979a43d4df8e5453c52ca607b77d364d64", "generation_seconds_batch": 17.893584094941616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.042638778686523, "target_z": 1.8359375648898355, "ce": 6.038498878479004, "genuineness": 4.093128204345703, "identity": 0.6048453450202942, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/552cac04dd072abc56293fd14a957d8442ba02625d4f45d4ccf4afb26afb38f6.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "560367cbe0f64e867e35c39b56ee57f80fad9d4d2bd1b732d238c0ffc5876074", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff69963efc0a96cb0eb8d484bf03e7af2df735fb527ba85ad36313b1263c06cf", "generation_seconds_batch": 24.47086635499727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25568151473999, "target_z": 1.9843749428969448, "ce": 6.229123115539551, "genuineness": 3.47493839263916, "identity": 0.1513315737247467, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/560367cbe0f64e867e35c39b56ee57f80fad9d4d2bd1b732d238c0ffc5876074.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "56110898ab2a84d9442d3e8918d3792b914378683149388cbdb80be1860f9b70", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "ce5464fe9594a0692fa791af19bae68a5e5fca05666d16230570e34c741e8203", "generation_seconds_batch": 18.12809932802338, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4483611583709717, "target_z": 1.4218750467206815, "ce": 6.205295085906982, "genuineness": 3.9173994064331055, "identity": 0.3358853757381439, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/56110898ab2a84d9442d3e8918d3792b914378683149388cbdb80be1860f9b70.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": "562a6ee9a7567cc6401c35d78e9273ac43a5e2debb21d551a1ec89a8091266eb", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The keys are on the kitchen table.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1901c73ff2b0bcaa0059ff585aa492724ce988a854a331a698097651250cfdaf", "generation_seconds_batch": 17.58346256799996, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 6.3641486167907715, "genuineness": 4.13078498840332, "identity": 0.45342838764190674, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/562a6ee9a7567cc6401c35d78e9273ac43a5e2debb21d551a1ec89a8091266eb.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "56b6568c3af2272c8e13a9adafed49e4c71deef258c45091077bf14bcfc3b1ce", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "582c1db45aea01eb3cc65e57f6d69d4ccd2ba34a97f275a8c7d4bebe8b1512ae", "generation_seconds_batch": 22.238272781018168, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9886369705200195, "target_z": 1.1015625389339012, "ce": 5.6254801750183105, "genuineness": 3.9550557136535645, "identity": 0.17321987450122833, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/56b6568c3af2272c8e13a9adafed49e4c71deef258c45091077bf14bcfc3b1ce.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": "5718c47e6725682cf43f50f63d8f7305796417f51df9b4354144ce7f402e6373", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d7d96e7c1d428da23fbe4153b455beeb5c6c51fd54bfcc53b4e4da48dbd79a3e", "generation_seconds_batch": 26.32801786600612, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.143553733825684, "target_z": 1.9062499896176264, "ce": 6.173524379730225, "genuineness": 4.256305694580078, "identity": 0.7291601896286011, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5718c47e6725682cf43f50f63d8f7305796417f51df9b4354144ce7f402e6373.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "572a659b5a4e5b6170a64e6ea93f0bdf05f3fa63ef639dd1f9d3e99f9d0be3f6", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "983b1884e288388eb6327a9585e75b5c896519828c07c0c2382044ee6485ae32", "generation_seconds_batch": 15.23614059900865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.098702430725098, "target_z": 1.8749998754115158, "ce": 6.157639026641846, "genuineness": 4.118232727050781, "identity": 0.48678863048553467, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/572a659b5a4e5b6170a64e6ea93f0bdf05f3fa63ef639dd1f9d3e99f9d0be3f6.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "57cfd5af82873d574eb7255198f2e60d00f2dcf4065c330ddfe9ee01d3bd1b82", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "88cb82575fb11b83a9e89c1f932b26f3608ebb9354caf7561d43c9e0001d32df", "generation_seconds_batch": 24.69295376900118, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.046441555023193, "genuineness": 4.319066047668457, "identity": 0.6610010862350464, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/57cfd5af82873d574eb7255198f2e60d00f2dcf4065c330ddfe9ee01d3bd1b82.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "57d8d10992330214327386d2d453011dfdddd3085eafe39d3cdf9a68c8866784", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2343342824694f5df808ae201293373588c6a41baeea3c83be335abc3382aa1f", "generation_seconds_batch": 18.28934576001484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.59206485748291, "target_z": 2.2187498027348997, "ce": 6.284721851348877, "genuineness": 4.250029563903809, "identity": 0.5008673667907715, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/57d8d10992330214327386d2d453011dfdddd3085eafe39d3cdf9a68c8866784.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": "5840a465afab871ffe456481eff5e485e759e8ce587718062febf7c150434994", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Someone moved the little blue chair.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e78994600f840ed9aac81f1ba067a102813f4346aa3960cdbc58e1d66c3ab38a", "generation_seconds_batch": 22.933450629003346, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.031425952911377, "target_z": 1.828125036338308, "ce": 6.348263263702393, "genuineness": 4.15588903427124, "identity": 0.45405113697052, "hypothesis": "Someone moved the little blue chair. It used to stand here beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5840a465afab871ffe456481eff5e485e759e8ce587718062febf7c150434994.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "5866ae9eb9b04d77b6baea7741dc541b33da96ac2146ae70ed67ded1e3edd499", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "4b0ea0679a7a97557aa2321eeabc382f9199676e4ecb909a98700fb8e47325dd", "generation_seconds_batch": 16.489074416924268, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.999849796295166, "target_z": 1.109375067485429, "ce": 5.855817794799805, "genuineness": 4.143336772918701, "identity": 0.10558412969112396, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5866ae9eb9b04d77b6baea7741dc541b33da96ac2146ae70ed67ded1e3edd499.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "590f9faa7580bda7b63db028faeefc3d8dd1cca315fec35fea0350d71dd069ea", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f041bbb18c525a61ffcc10f14896bd6bd0753ce277dccd675832e0c764b2acb", "generation_seconds_batch": 24.736399626010098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.199617385864258, "target_z": 1.9453123001393064, "ce": 6.133810997009277, "genuineness": 4.149612903594971, "identity": 0.8014358282089233, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/590f9faa7580bda7b63db028faeefc3d8dd1cca315fec35fea0350d71dd069ea.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "59d93d48625ad2f8b2e9170ebdb39d084f02ebe4473f691aa1e834cec4158c8a", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "d123c8d75cfa0636ac847fb09f27d2a8978c793790b453a43cc4bfa5bfe5a6ca", "generation_seconds_batch": 18.018651907914318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 6.1020402908325195, "genuineness": 4.268857479095459, "identity": 0.6447352170944214, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/59d93d48625ad2f8b2e9170ebdb39d084f02ebe4473f691aa1e834cec4158c8a.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "5a06bc3105c10a3dd1883aefebb3d8d7a6fc3b5f67e713c41b507c119631dd92", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "fdd603277f85328a8980bbaca320ac4e179b2a321436b6ca0d264674840eb3a6", "generation_seconds_batch": 28.242928871070035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8295958042144775, "target_z": 1.6875000207647475, "ce": 6.149696350097656, "genuineness": 3.9048471450805664, "identity": 0.6645660400390625, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5a06bc3105c10a3dd1883aefebb3d8d7a6fc3b5f67e713c41b507c119631dd92.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "5a1922735349b2c5f8218f8df9694f8dbfff603d076bd9a21e0059279f7cf0e8", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "5e8db7004a1345174e205eecb12ea797e49427e8d628b6f8044dbdedefb11023", "generation_seconds_batch": 24.12391836300958, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1884050369262695, "target_z": 1.937500103823737, "ce": 6.24500846862793, "genuineness": 3.540836811065674, "identity": 0.39675676822662354, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5a1922735349b2c5f8218f8df9694f8dbfff603d076bd9a21e0059279f7cf0e8.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": "5a2ceaa6cbd0032e703918248abd6dd9c235b8155a5f03cdf3d9d74b120b5eb1", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "446dcb070eb96abb67fcfe572d56a9e71daf384fcdecb82fa46cbaf4484ba1fd", "generation_seconds_batch": 21.32970803801436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8295958042144775, "target_z": 1.6875000207647475, "ce": 6.268836498260498, "genuineness": 4.143336772918701, "identity": 0.6001851558685303, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5a2ceaa6cbd0032e703918248abd6dd9c235b8155a5f03cdf3d9d74b120b5eb1.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "5a3f022695f629743fdf7454e8b04c5294da1ccfd6ea8868bd8276750172d41c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9c570139faeb00b0d4664834750443cc8eee852778cae74bf08b1defd8bb31dd", "generation_seconds_batch": 24.068633335991763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.042638778686523, "target_z": 1.8359375648898355, "ce": 5.831989288330078, "genuineness": 3.9676074981689453, "identity": 0.533014178276062, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay de.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5a3f022695f629743fdf7454e8b04c5294da1ccfd6ea8868bd8276750172d41c.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "5ad99584bed6590df7a84cb5d42a82e59068f8999b41ed7f6e08560ad22daa2c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "823759ac20b4711258c4389a8b088768502c895aed84f8dd4e0ede62ecf7363e", "generation_seconds_batch": 24.0181316989474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7847447395324707, "target_z": 1.6562500726766158, "ce": 6.133810997009277, "genuineness": 4.375550270080566, "identity": 0.7348436117172241, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5ad99584bed6590df7a84cb5d42a82e59068f8999b41ed7f6e08560ad22daa2c.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": "5b37024aff55aa4cca40c33d2300fe29af0a6d344cd54c5b8873a0967b90cdb6", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2fe48ef30e73537257c55bf10a90ca8ae1947d5a94d5167055d85d53440c9429", "generation_seconds_batch": 22.95962944708299, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.605340003967285, "target_z": 1.5312499480881316, "ce": 6.308549880981445, "genuineness": 3.9048471450805664, "identity": 0.6706202626228333, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5b37024aff55aa4cca40c33d2300fe29af0a6d344cd54c5b8873a0967b90cdb6.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "5b37b55580c3a8c5d6445febdcd52e72a699f93bc531751c147f7cbaeec30df6", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "fd156dd8b822c2eae34de9cc8f7fd94e4abc7d5a63d49461333c0bc218d356af", "generation_seconds_batch": 24.270309402025305, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7623190879821777, "target_z": 1.6406250155735604, "ce": 6.252951145172119, "genuineness": 4.093128204345703, "identity": 0.6493600606918335, "hypothesis": "Komt u's, ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.1111111111111111, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5b37b55580c3a8c5d6445febdcd52e72a699f93bc531751c147f7cbaeec30df6.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "5b872ac4202acdc651b2ee3cb0b6a5be6958c265b9cc32ba16726d9337f8a7c4", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4ff8ae9823b88cc1706abbcc6d0aefccc4d5acaa63a4af3bbee0916018d2531f", "generation_seconds_batch": 23.70501429901924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.569639682769775, "target_z": 2.2031250778678024, "ce": 6.4912309646606445, "genuineness": 3.8169827461242676, "identity": 0.586301326751709, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5b872ac4202acdc651b2ee3cb0b6a5be6958c265b9cc32ba16726d9337f8a7c4.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "5bb8b38aa0b90bd960366e2092f00937e30cee1578bf2e3faf9dce5422c50de9", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e3251ec234da59a23b74e4e72b651516e65a811c2c84212d841a1aba2204aa9c", "generation_seconds_batch": 17.796968202921562, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9753620624542236, "target_z": 1.7890625596986487, "ce": 5.732706069946289, "genuineness": 3.7291181087493896, "identity": 0.2744135856628418, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5bb8b38aa0b90bd960366e2092f00937e30cee1578bf2e3faf9dce5422c50de9.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "5bee30f82a018aa59ca370365bfb7f7c0df99f84138049c76a00009627b7c0b8", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "346bc604dc2294c1a54444d1c5b56b043aa359e5300828d891ef28ca3bb9291c", "generation_seconds_batch": 23.152368337963708, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.627765655517578, "target_z": 1.5468750051911868, "ce": 6.046441555023193, "genuineness": 4.017816066741943, "identity": 0.7265916466712952, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5bee30f82a018aa59ca370365bfb7f7c0df99f84138049c76a00009627b7c0b8.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "5c831dfb9a0070ac4daa174ee19a21da63859fd6b0b8385dd8519eea93fecc2b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "88fcc3022e41c2fc2f29dcb9d2c34f88f1abeead19559e407a3495c4a02687e2", "generation_seconds_batch": 24.38863174605649, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.098702430725098, "target_z": 1.8749998754115158, "ce": 6.062326908111572, "genuineness": 4.061748504638672, "identity": 0.6869018077850342, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5c831dfb9a0070ac4daa174ee19a21da63859fd6b0b8385dd8519eea93fecc2b.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "5cf314572c398dee9bf219caa4042e160bad56d858ad787a439e951bcbd09aad", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "72c5837e966cfc1727cc121530a4ebe43248d5baff477410ae56535c5e08a839", "generation_seconds_batch": 17.51401129900478, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.358658790588379, "target_z": 1.3593749844264396, "ce": 6.650084495544434, "genuineness": 4.15588903427124, "identity": 0.1093064695596695, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5cf314572c398dee9bf219caa4042e160bad56d858ad787a439e951bcbd09aad.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": "5d0b6f57a92146b7837141ef573ce4a35063aea369b93eb7ae80d9664c41c04f", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d20e9d1d0736eb37f8fa36590a89630f1a3dd8b8791cf7bfd94e5bb5a5527091", "generation_seconds_batch": 18.307272696984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25568151473999, "target_z": 1.9843749428969448, "ce": 6.109982967376709, "genuineness": 4.143336772918701, "identity": 0.6988735198974609, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5d0b6f57a92146b7837141ef573ce4a35063aea369b93eb7ae80d9664c41c04f.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "5d63006b8d40a37f7bb2d9837b676844658f160aaccac74566053fefab0f4f83", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "db3fa2067aa0f96fd4ddd5788cc5500b8d697f2d68456d08e9ec48ae0639963c", "generation_seconds_batch": 17.47214636905119, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.222043037414551, "target_z": 1.9609373572423618, "ce": 6.133810997009277, "genuineness": 3.9801597595214844, "identity": 0.7934836745262146, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5d63006b8d40a37f7bb2d9837b676844658f160aaccac74566053fefab0f4f83.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "5de335ce8b9a11b85df0f5df78e54d7b7b9491d380036079182b6c0c3720c209", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2949f1c16cbd6a8c8a50927771df8bf531cd6e1f437b7dbd7c6925fc9c10b175", "generation_seconds_batch": 24.881046845926903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7398934364318848, "target_z": 1.6249999584705053, "ce": 6.157639026641846, "genuineness": 4.268857479095459, "identity": 0.6758983731269836, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5de335ce8b9a11b85df0f5df78e54d7b7b9491d380036079182b6c0c3720c209.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "5e06fc6e5650e0cba96c9f84f20682386ca693af6dd362a5e5d15b180d4ccb91", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "cf6681e2612ddbbee4ec8ed12119d5879782e2e30e915daee8b88b73dfd1be36", "generation_seconds_batch": 18.289270548033528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.964149236679077, "target_z": 1.781250031147121, "ce": 6.332377910614014, "genuineness": 4.293961524963379, "identity": 0.6710375547409058, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5e06fc6e5650e0cba96c9f84f20682386ca693af6dd362a5e5d15b180d4ccb91.mp3" }, { "id": "heldout-fr-2", "language": "fr", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-2", "key": "5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b71c7a3450189eaafd4fff3b9181339678d16ee26f103ab6b1682c4cd4de84d", "generation_seconds_batch": 17.17794508393854, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9192981719970703, "target_z": 1.7500000830589895, "ce": 6.054384231567383, "genuineness": 4.281409740447998, "identity": 0.7112889289855957, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "5e627acbc4a628ac219333677fc8e7c94c2c8fbc9a52ab5c4b1330c93d6cbeb6", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "c3d7d71ea833173c5468fe5c03085b5605ba06853f104c1e9d81ea478439798a", "generation_seconds_batch": 25.022201505023986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.457511901855469, "target_z": 2.1250001245884844, "ce": 6.141753673553467, "genuineness": 4.093128204345703, "identity": 0.5713295936584473, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5e627acbc4a628ac219333677fc8e7c94c2c8fbc9a52ab5c4b1330c93d6cbeb6.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "5f0f372637592f0c043ba17128b9007febc1ff679479ebf4bfcd36169314aeb8", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "6a4736fa66aab9678ef9f2e3e0a900bc2d3c444bf18d4914d207e4a87593402f", "generation_seconds_batch": 23.933880429016426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.031425952911377, "target_z": 1.828125036338308, "ce": 6.133810997009277, "genuineness": 4.042920112609863, "identity": 0.6680207848548889, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5f0f372637592f0c043ba17128b9007febc1ff679479ebf4bfcd36169314aeb8.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": "5f1ca710a39c3e4ffa00c711e45d0c1084281ff5e6026d7729b65e28fd8ad81d", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7e03d7ecfe463c547e7594f723c349c99cd8ecec09955ed1fb2a1695fec97ef5", "generation_seconds_batch": 15.633880621986464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.502363204956055, "target_z": 2.1562502387945948, "ce": 6.292664527893066, "genuineness": 4.093128204345703, "identity": 0.5997740626335144, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5f1ca710a39c3e4ffa00c711e45d0c1084281ff5e6026d7729b65e28fd8ad81d.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "5f7d3ef323dd3f294bdb24093097c816e181d532faa13c606302e8d4f3033e0c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "c24ff9752b9fce77820a2440ac74333e5f9c86877fa69bbe1646f58689930c2f", "generation_seconds_batch": 32.64235888503026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.627765655517578, "target_z": 1.5468750051911868, "ce": 5.498397350311279, "genuineness": 3.848362922668457, "identity": 0.47785818576812744, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5f7d3ef323dd3f294bdb24093097c816e181d532faa13c606302e8d4f3033e0c.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "5fb9a2a8e0dfd845598f51bf2926e075cce90ba58eb975eb4d20a0a5bcb6b49d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "041adff814c54a0c2a0e4601465488f931e8f748cb8b041c30b8df1dc0e4748e", "generation_seconds_batch": 21.535385253024288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.15476655960083, "target_z": 1.914062518169154, "ce": 5.831989288330078, "genuineness": 3.8044304847717285, "identity": 0.41734790802001953, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/5fb9a2a8e0dfd845598f51bf2926e075cce90ba58eb975eb4d20a0a5bcb6b49d.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "6008931961a3208c62f8957bb79735f602fa09cff280fb687b902c690272414c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "5a8c1c9835904b05bd66a314af484ca1fcaa821552402f028031cf1f429db636", "generation_seconds_batch": 24.369237206992693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.636916160583496, "target_z": 2.2499999169410105, "ce": 6.403861999511719, "genuineness": 4.344170093536377, "identity": 0.3927341401576996, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6008931961a3208c62f8957bb79735f602fa09cff280fb687b902c690272414c.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "606dd6056f04e853d2cc584b465c5b984b49915db5b7e7dfe86f66aa8ddcad69", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b64056f6a12443be45c4f754b7aa4a8801550170a5597b4a3ddaa3265805cf6", "generation_seconds_batch": 24.274982999893837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.435086250305176, "target_z": 2.109375067485429, "ce": 6.3641486167907715, "genuineness": 3.6663577556610107, "identity": 0.3057708144187927, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/606dd6056f04e853d2cc584b465c5b984b49915db5b7e7dfe86f66aa8ddcad69.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "60d56be7a3cf195771fdbcb686b264fc0096f700d8364111f3f6bd4d344a35c1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db608b82010fe3fab402eb96fa7d3959aa4e6007948dbdd69bb796f0bf8dc928", "generation_seconds_batch": 24.44029906403739, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.908085346221924, "target_z": 1.7421875545074619, "ce": 6.0146708488464355, "genuineness": 4.055472373962402, "identity": 0.59614098072052, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/60d56be7a3cf195771fdbcb686b264fc0096f700d8364111f3f6bd4d344a35c1.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "60ef68ee579996074e898fab53dcb4093ab792fbbcf0c3933cc4e8c3b9aaab45", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "b2b5a2821912c88d900b7acb92b3e2dcf5c994ddc69b0bc2aa646e8461364269", "generation_seconds_batch": 23.828927029971965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.278107166290283, "target_z": 2.0, "ce": 6.038498878479004, "genuineness": 4.105680465698242, "identity": 0.6406967639923096, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/60ef68ee579996074e898fab53dcb4093ab792fbbcf0c3933cc4e8c3b9aaab45.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": "60f4940ad30fb7865e2d0503ed7cc10bd99fd95c561ed3631f303ef33555d4d9", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c03116f0c3bf7926871f4df825934bf5f3713974d24afcf0b1233d456849452b", "generation_seconds_batch": 21.010833706939593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.840808629989624, "target_z": 1.695312549316275, "ce": 6.046441555023193, "genuineness": 4.055472373962402, "identity": 0.6888827681541443, "hypothesis": "De sleutels liggen op de keukentafel. Ik ik in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/60f4940ad30fb7865e2d0503ed7cc10bd99fd95c561ed3631f303ef33555d4d9.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "61c3d49ac4043aa736aad79c7f18ecf75c449badba75a039f18a34080f479f2e", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "1f47c3a1584082c2eb74a8c4b066c24b39c730d578cfe7e4ab63a78c360ffb86", "generation_seconds_batch": 26.436338186031207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.908085346221924, "target_z": 1.7421875545074619, "ce": 6.252951145172119, "genuineness": 4.369274139404297, "identity": 0.7609498500823975, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/61c3d49ac4043aa736aad79c7f18ecf75c449badba75a039f18a34080f479f2e.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "61de6a934b86a7402760ef80b11e07f3101190452e9acb5bcca4513f69479b74", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ede55b0f3e960a3a07a5accdc5526b983e1be401d110953d186a117d34388fa9", "generation_seconds_batch": 17.97186829696875, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7062551975250244, "target_z": 1.6015625389339012, "ce": 5.657250881195068, "genuineness": 3.9299514293670654, "identity": 0.6119226813316345, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/61de6a934b86a7402760ef80b11e07f3101190452e9acb5bcca4513f69479b74.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "61ebfcf9c764f30bce8c6f22569bd9bf3927a5fe13477d95502c8974e48166b6", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "50d9eeca4894858d13bf78b33e7b4cbf0e6eab922dc6787afdb3c847ec9d39fa", "generation_seconds_batch": 23.161296072998084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.695042371749878, "target_z": 1.5937500103823736, "ce": 6.213237762451172, "genuineness": 3.9299514293670654, "identity": 0.6208270192146301, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/61ebfcf9c764f30bce8c6f22569bd9bf3927a5fe13477d95502c8974e48166b6.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": "62cf6a08a4edc3180c61fce69c62223e8d36794bdb2136a0c9c58b46fae58c3e", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "472a19f6aaaad1b650608bf0456783f8c46380abe9eae6251d7435ba079e2d38", "generation_seconds_batch": 21.32970803801436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.986574649810791, "target_z": 1.7968749221321973, "ce": 6.197352409362793, "genuineness": 4.080576419830322, "identity": 0.46470558643341064, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/62cf6a08a4edc3180c61fce69c62223e8d36794bdb2136a0c9c58b46fae58c3e.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "632ca293ee1c906c3dd05a2c5c530bf7b55a0e39b7f23727224796d96e04f4d1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "12ed63cd0a1ad7dffb6cf5a7e00f5ade895b0e2abca72c3b86cefe8fd2993173", "generation_seconds_batch": 18.212289302959107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0202131271362305, "target_z": 1.8203125077867803, "ce": 6.1894097328186035, "genuineness": 4.149612903594971, "identity": 0.5132704973220825, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/632ca293ee1c906c3dd05a2c5c530bf7b55a0e39b7f23727224796d96e04f4d1.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "6364b5c030c7567398cc03e177e79744e3b9692ae291cc071092bab494b59491", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "5aa8c8afeaaf4a3e104196454aedaa62a45a97ffe2d64eb7cbddd5b8da93d694", "generation_seconds_batch": 25.715329775004648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6165528297424316, "target_z": 1.5390624766396592, "ce": 6.260893821716309, "genuineness": 4.218649387359619, "identity": 0.7232096195220947, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6364b5c030c7567398cc03e177e79744e3b9692ae291cc071092bab494b59491.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "636e4656fae5a21134c90c7f63b41cc0855ea04074d09e1c36c2d86e4d487b96", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "1845ec034233dd269e39ce49c2bda7d28dab6c52bbbc9ab0bc43141492029c6a", "generation_seconds_batch": 23.163789084064774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 6.086154937744141, "genuineness": 3.7165660858154297, "identity": 0.41769057512283325, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/636e4656fae5a21134c90c7f63b41cc0855ea04074d09e1c36c2d86e4d487b96.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "63deb8cd78219beb794d3fac42693705e7f7ee555307b05af5dc2fd0907e04a5", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "db8929a66c3d7b784c5c4aa96dbcaeac775cd3367a0f8a5626be4b5002f4074b", "generation_seconds_batch": 18.006340182968415, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.031425952911377, "target_z": 1.828125036338308, "ce": 5.927301406860352, "genuineness": 3.9676074981689453, "identity": 0.4612027406692505, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo para puerta.", "wer": 0.13333333333333333, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/63deb8cd78219beb794d3fac42693705e7f7ee555307b05af5dc2fd0907e04a5.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": "640829a6ce4d9a1e3affcc96d033203d82d5ac4f28f3667739e1c8bae7f01b52", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f84304a863914c93783dfb16ac13f9bdc54eea909bfa609cec24397f2c5b5567", "generation_seconds_batch": 18.15453819592949, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.479937553405762, "target_z": 2.1406251816915396, "ce": 6.205295085906982, "genuineness": 4.13078498840332, "identity": 0.5831871032714844, "hypothesis": "La carretera gira a la izquierda, tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/640829a6ce4d9a1e3affcc96d033203d82d5ac4f28f3667739e1c8bae7f01b52.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "65370ace4e3b8361eb36987d53331b4ede56bebb886c515612ecc16a520ce621", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "2dc624a86a8fa16ed26de96c8c13a8b196592c5fe5e9826d09bd33f3bda08c19", "generation_seconds_batch": 23.79455260199029, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0202131271362305, "target_z": 1.8203125077867803, "ce": 6.324435234069824, "genuineness": 3.9550557136535645, "identity": 0.416189044713974, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/65370ace4e3b8361eb36987d53331b4ede56bebb886c515612ecc16a520ce621.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "656f7cf0feff1adcefe8235e8a350eb0e8d8390618c8a7afdead1052827b13dd", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "55337e51fe8f7f2c50a3533b361238fa374fb291300baf17ab000b6043664686", "generation_seconds_batch": 26.781639675959013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6389784812927246, "target_z": 1.5546875337427144, "ce": 6.157639026641846, "genuineness": 4.005263805389404, "identity": 0.6153831481933594, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/656f7cf0feff1adcefe8235e8a350eb0e8d8390618c8a7afdead1052827b13dd.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "6581329a2f9dc5fd0fa1977f61e9fa45b352f32df367c5ea931c40dc4b6b7249", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6fb88253c15977ed6594ecf99a37f5af5e8d1f3d785c7b973dd9782444eda19e", "generation_seconds_batch": 24.552437983918935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2801694869995117, "target_z": 1.3046876168017039, "ce": 6.1020402908325195, "genuineness": 3.9927117824554443, "identity": 0.3477923274040222, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6581329a2f9dc5fd0fa1977f61e9fa45b352f32df367c5ea931c40dc4b6b7249.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "6646ed704c56e709aef1e91f93b678b0ff5f17c88b95a35ed68363bd7e8af307", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9b59db7a47584394636d988e596767fdece82562f8b9c09aa0f1c16d5c4e25e5", "generation_seconds_batch": 19.37459076696541, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.098702430725098, "target_z": 1.8749998754115158, "ce": 6.181467056274414, "genuineness": 3.9927117824554443, "identity": 0.6900726556777954, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6646ed704c56e709aef1e91f93b678b0ff5f17c88b95a35ed68363bd7e8af307.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "66b8a6f864bee88d72656d4324765cdc5ebf929b6eefb028ee4d9daebb5e2d96", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "d291446a5135626ed1c88d8245f4a7dad463be0bb15f1dd6acecd19e4f464a3b", "generation_seconds_batch": 25.312129957950674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.861171722412109, "target_z": 2.4062498234996474, "ce": 6.181467056274414, "genuineness": 4.231201171875, "identity": 0.48673015832901, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/66b8a6f864bee88d72656d4324765cdc5ebf929b6eefb028ee4d9daebb5e2d96.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "673c40ef2d27348e2b9221e4ccc2101d9c00257cebcd471dc560cf5cc9793c78", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d43af0998bdf16add6dcce4d2735522db46f48d343ff6a8c4ecb2e90b68ce0e1", "generation_seconds_batch": 22.92360788001679, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8071701526641846, "target_z": 1.671874963661692, "ce": 6.141753673553467, "genuineness": 4.432034492492676, "identity": 0.4538534879684448, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/673c40ef2d27348e2b9221e4ccc2101d9c00257cebcd471dc560cf5cc9793c78.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "6788dac46c4e4a1c79257cc66ad2452192a68dfd41ccda455d75ca2ea23f5763", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "3dfc660714041db9500163cc5ec391a61a43c7e2188cbdfa5d628966831d51f8", "generation_seconds_batch": 27.688737157033756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.266894340515137, "target_z": 1.9921874714484724, "ce": 6.23706579208374, "genuineness": 4.042920112609863, "identity": 0.3525756001472473, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6788dac46c4e4a1c79257cc66ad2452192a68dfd41ccda455d75ca2ea23f5763.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "6793205a775e8cf2cdb55bdd839a4a6c1130bcb0bc751a9efa286dfe27466bcc", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "293a953a171d1a66175b8a52f1baaf8fb67ff42fd3e341fa1c39b22a7b596cce", "generation_seconds_batch": 20.876710272976197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8744468688964844, "target_z": 1.718749968852879, "ce": 5.982900142669678, "genuineness": 4.375550270080566, "identity": 0.48241761326789856, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6793205a775e8cf2cdb55bdd839a4a6c1130bcb0bc751a9efa286dfe27466bcc.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "680ec35c75c1fbc05112303dbf59d3186fb929cf89b5b5871219bae69f14f31d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "b0d849c32d269a93011bfcdee18113a345e86e730ddd7a0775a00d4214195d40", "generation_seconds_batch": 23.183536314056255, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.840808629989624, "target_z": 1.695312549316275, "ce": 6.205295085906982, "genuineness": 4.2249250411987305, "identity": 0.4166245460510254, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/680ec35c75c1fbc05112303dbf59d3186fb929cf89b5b5871219bae69f14f31d.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "6a4dbcc7cbb0a3929fce4e519ec430d988ead1886c77dedd4800c90d01ad167f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "6838e3e91bb596687b7fbbdba2c380fe7beec0c25b3a60b96f17a0a6c2c7ad5b", "generation_seconds_batch": 23.191972568980418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.121128082275391, "target_z": 1.890624932514571, "ce": 6.149696350097656, "genuineness": 4.218649387359619, "identity": 0.6224275231361389, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6a4dbcc7cbb0a3929fce4e519ec430d988ead1886c77dedd4800c90d01ad167f.mp3" }, { "id": "heldout-en-4", "language": "en", "sentences": [ "I wrote the number on the envelope.", "Read it carefully... and check the last two digits.", "Call me when you have a moment." ], "text": "I wrote the number on the envelope. Read it carefully... and check the last two digits. Call me when you have a moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-4", "key": "6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca6739707c20607d44663a1a2aa493e1df9a25141dd11f87311e0653f0df42e1", "generation_seconds_batch": 17.392608234891668, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5380632877349854, "target_z": 1.4843749428969448, "ce": 6.2767791748046875, "genuineness": 3.948779582977295, "identity": 0.39870813488960266, "hypothesis": "I wrote the number on the envelope, read it carefully, and checked the last two digits. Call me when you have a moment.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "6b0071c1cd773df2e92e53883d271bd4a99357b835130c7c6f7289f73a20e1d1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "372d8efa848e8a0c894b94f3ec0e8d0c4db081f350e01b2d188d365b409f5686", "generation_seconds_batch": 17.182650041067973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7623190879821777, "target_z": 1.6406250155735604, "ce": 5.903473377227783, "genuineness": 3.886019229888916, "identity": 0.7834653854370117, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6b0071c1cd773df2e92e53883d271bd4a99357b835130c7c6f7289f73a20e1d1.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c721ddd5e1f8f2c96c50011b8743e79c0a520cb30397539e194113b30e18513", "generation_seconds_batch": 20.281209568027407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.145616054534912, "target_z": 1.2109376064193302, "ce": 6.23706579208374, "genuineness": 4.6956281661987305, "identity": 0.6921969056129456, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "6bdfe0120be48cccf815018fb75f73be473e07bb752c9542175d5c6edc0f790e", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "7a27d8726fac8bd0021f2d97ee80635671ff543b7df9894658e12f9a4e11781d", "generation_seconds_batch": 24.042246051016264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.840808629989624, "target_z": 1.695312549316275, "ce": 5.879645347595215, "genuineness": 3.798154592514038, "identity": 0.6359407901763916, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6bdfe0120be48cccf815018fb75f73be473e07bb752c9542175d5c6edc0f790e.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": "6c089ceabfe150049d9a2727a16678e650728852af9c5732ecf97161d5c9050d", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The road turns left behind the bridge.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dcfe9d08b80c82a747f560cfb1b7cfa997026e80da6c624f58c190af5068caa0", "generation_seconds_batch": 22.960852851043455, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5604889392852783, "target_z": 1.5, "ce": 6.41180419921875, "genuineness": 4.024092197418213, "identity": 0.3175700306892395, "hypothesis": "The road turns left behind the bridge. Keep going. Until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6c089ceabfe150049d9a2727a16678e650728852af9c5732ecf97161d5c9050d.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "6d394a689c128ca1f181685e31fc466d27750b6fa8843a7604fab38e9b0fd3ab", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "b8038135cb4f18d208300e953fae0346a3c33cf755859ac7c0b0a7b0560c9a13", "generation_seconds_batch": 23.314095688052475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1568286418914795, "target_z": 1.218749968852879, "ce": 5.895530700683594, "genuineness": 3.9173994064331055, "identity": 0.551737904548645, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6d394a689c128ca1f181685e31fc466d27750b6fa8843a7604fab38e9b0fd3ab.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "6d42c055ded8fe159ab6114e2875685a40a76ba3ed2741223ec8d50a2981ce95", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "9eddc7d273d449c29682624b8c47ef8fad691d5b779c4f84414418f5b3357898", "generation_seconds_batch": 22.561396892997436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4259355068206787, "target_z": 1.4062499896176264, "ce": 6.300607204437256, "genuineness": 4.300237655639648, "identity": 0.5079613924026489, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6d42c055ded8fe159ab6114e2875685a40a76ba3ed2741223ec8d50a2981ce95.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": "6d5499fc42f8437538b465ecbabc73c622f4e18b347d8fe29a70fb88560daa36", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "84f615a705f7776c94ba6f028d00367e226a0a327463c3f051cdbe1c431305d8", "generation_seconds_batch": 23.78272386395838, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.547214031219482, "target_z": 2.1875000207647473, "ce": 6.086154937744141, "genuineness": 4.168440818786621, "identity": 0.7726315259933472, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lis le bien et vérifie les deux derniers chiffres. Appelle moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6d5499fc42f8437538b465ecbabc73c622f4e18b347d8fe29a70fb88560daa36.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "6dcda23d944705b3ea22acc1e3fe37fd7d680e4742be0a42898246e81922cfe1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "eb38685e203d90996bd81c3be6cde40e240e91811b825fe0f3ec604d5e3b9bb4", "generation_seconds_batch": 20.3937280169921, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.547214031219482, "target_z": 2.1875000207647473, "ce": 6.292664527893066, "genuineness": 3.998987913131714, "identity": 0.6879075765609741, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6dcda23d944705b3ea22acc1e3fe37fd7d680e4742be0a42898246e81922cfe1.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "6e3327d90a9669238bb8f65bec11f8667dc97fa4a084a5216b2e5e65c8635300", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "42816575069bc3c2665a2371bb8f92bc2a092767f08b81e6d2b17eb427fcb5be", "generation_seconds_batch": 27.829898354946636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4932122230529785, "target_z": 1.4531249948088132, "ce": 6.24500846862793, "genuineness": 4.2374773025512695, "identity": 0.2955023944377899, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6e3327d90a9669238bb8f65bec11f8667dc97fa4a084a5216b2e5e65c8635300.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "6e87d07b844de9517b1eb08b8b1315d4a685f1dd8092f9fd95916d8e22822b4b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "b0d2c41338f6e7fbae82e50f149e5daeab7b5c276896d6df75591e2803fd0791", "generation_seconds_batch": 24.317695157951675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2108306884765625, "target_z": 1.9531251609267921, "ce": 5.990842819213867, "genuineness": 4.055472373962402, "identity": 0.3739944398403168, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6e87d07b844de9517b1eb08b8b1315d4a685f1dd8092f9fd95916d8e22822b4b.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "6eba949d8337aa6126d47d8d2185826560d6bffc40c3ff835e2c0031437c9265", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a128a67aa99ab8739242201441803683049f04bdd460a34c08ec91b58462525b", "generation_seconds_batch": 24.15808807394933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.435086250305176, "target_z": 2.109375067485429, "ce": 6.419747352600098, "genuineness": 4.444586753845215, "identity": 0.6457500457763672, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6eba949d8337aa6126d47d8d2185826560d6bffc40c3ff835e2c0031437c9265.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "6f8d61260565ad3c5931b301c6ae2992d8bbcae523ab9d6d9937c1f4fb6d215b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "d93c4a3c70b0121a879f10829af0db61459934b575286a2f8fc78133859f4095", "generation_seconds_batch": 22.363289171946235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5380632877349854, "target_z": 1.4843749428969448, "ce": 6.268836498260498, "genuineness": 4.011539936065674, "identity": 0.7875449657440186, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6f8d61260565ad3c5931b301c6ae2992d8bbcae523ab9d6d9937c1f4fb6d215b.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "6fc855a1f310aff586d6a45c90e45113a4f4c0a6139ae7b3920b9e4626ec827f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2d1fa23edff7fa1bbcb3c4ec8c397b97ee46074b63a2a24d5b3b374c3b8badcc", "generation_seconds_batch": 17.14099936396815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.291382074356079, "target_z": 1.3124999792352525, "ce": 6.1020402908325195, "genuineness": 3.8295347690582275, "identity": 0.17326024174690247, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/6fc855a1f310aff586d6a45c90e45113a4f4c0a6139ae7b3920b9e4626ec827f.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "700b7466e742a7ba1aaa954420986af37f2bcb998f44a6e6bc7f816c880bcd05", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "9be9eb3efc4e53b186ab9ddd40dbbd5f2cc344e7044f024117a5952ec930c995", "generation_seconds_batch": 24.384073166991584, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2801694869995117, "target_z": 1.3046876168017039, "ce": 6.292664527893066, "genuineness": 4.093128204345703, "identity": 0.2830311059951782, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/700b7466e742a7ba1aaa954420986af37f2bcb998f44a6e6bc7f816c880bcd05.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "7049c60eb0bbb301efd190cf4c823faf111c7c8213e51f2aa30593f49fd936b8", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "7dcb2569e6bb3c85afac622bfb40bc80b855f893c996714add3b803529d2c2a7", "generation_seconds_batch": 17.9515318099875, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.659341812133789, "target_z": 2.2656249740440657, "ce": 6.292664527893066, "genuineness": 4.13078498840332, "identity": 0.6108370423316956, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7049c60eb0bbb301efd190cf4c823faf111c7c8213e51f2aa30593f49fd936b8.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "71d46c77e5c9c221715b1d12ef921b019ad3f2bb3ac2a6e376a0683251e976c0", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "690e095f68a1de02b972433c049f5e92da18567e6e160a65a8bea9b8d79001d4", "generation_seconds_batch": 17.893584094941616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0202131271362305, "target_z": 1.8203125077867803, "ce": 5.550024509429932, "genuineness": 3.9299514293670654, "identity": 0.527727484703064, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/71d46c77e5c9c221715b1d12ef921b019ad3f2bb3ac2a6e376a0683251e976c0.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "720b014e4e03d718ad12b913a32e8fa4cd0e3a2c5fd6480db3a13678860651ba", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "b79b8576bc4d91a82d5c888b48fe2392d91d3a8ac1a95a4b1071b885bb607e64", "generation_seconds_batch": 22.92360788001679, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.098702430725098, "target_z": 1.8749998754115158, "ce": 6.229123115539551, "genuineness": 4.13078498840332, "identity": 0.722182035446167, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/720b014e4e03d718ad12b913a32e8fa4cd0e3a2c5fd6480db3a13678860651ba.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "724978fe7ac481dd19a03b22ad0f79095584baee43ae5309adcf31473a8523c9", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb4f60e6f0672bc5487508810856f960dde486b4b95f173db02ed8153b6cb7ee", "generation_seconds_batch": 23.342685795971192, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.614490509033203, "target_z": 2.2343748598379554, "ce": 6.09409761428833, "genuineness": 4.137060642242432, "identity": 0.608244776725769, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/724978fe7ac481dd19a03b22ad0f79095584baee43ae5309adcf31473a8523c9.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "72a6da16526dabf155a5e3b3b82a0648ffbd7c1cb738479e961cb0d0d4cd34ab", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "26bdb122a6bcd12143f89d50636218b5a896735b8e2da8a856f6b01825b879d7", "generation_seconds_batch": 24.23634832899552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9753620624542236, "target_z": 1.7890625596986487, "ce": 6.356205940246582, "genuineness": 4.293961524963379, "identity": 0.3406987190246582, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/72a6da16526dabf155a5e3b3b82a0648ffbd7c1cb738479e961cb0d0d4cd34ab.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "72bb4f54648d5ceb7abcc7003a884a43496b7060a2a0444a9a36d3541f1d2049", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5de54c0aab6b6d193f3a636cc85f2bc06c6aa21ef71b26ee8e1995f6dca8ce2", "generation_seconds_batch": 24.327050105086528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.065064430236816, "target_z": 1.851562621992891, "ce": 5.8875885009765625, "genuineness": 4.137060642242432, "identity": 0.5553451776504517, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/72bb4f54648d5ceb7abcc7003a884a43496b7060a2a0444a9a36d3541f1d2049.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "72d62e4d9a67f27749c298865a78e632f58aab35272ab97311d43b671167843f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2bdc382f8289a86a0163d5e3a07adae95639fe6757cd04b81188ebe94b895bad", "generation_seconds_batch": 24.31624027399812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.121128082275391, "target_z": 1.890624932514571, "ce": 5.824047088623047, "genuineness": 4.099404335021973, "identity": 0.6537127494812012, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/72d62e4d9a67f27749c298865a78e632f58aab35272ab97311d43b671167843f.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "72e5e2c174449f5899fcb25f6e0ec22aa73ea7df20738d8d16f29d3fa002c678", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1ecaa82aa8994f6d526307ec67f4094b8d95013ce2d5caa9be3c336fb3b961e4", "generation_seconds_batch": 25.42722071299795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.009000301361084, "target_z": 1.8124999792352525, "ce": 6.157639026641846, "genuineness": 4.055472373962402, "identity": 0.8240638971328735, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/72e5e2c174449f5899fcb25f6e0ec22aa73ea7df20738d8d16f29d3fa002c678.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "72f61036f36f2acecb5d711fe2f77a0ae0e9e4f018ea3f752a7b366213084af7", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fce95f1764272490e9a003fb4d71c2f1e2b8e764730dad6b2a45ae68ed0b06db", "generation_seconds_batch": 23.896377769997343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8520214557647705, "target_z": 1.7031250778678027, "ce": 5.6254801750183105, "genuineness": 4.005263805389404, "identity": 0.686766505241394, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/72f61036f36f2acecb5d711fe2f77a0ae0e9e4f018ea3f752a7b366213084af7.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "736d7ee32341a3c43f0d0ed6f92e4c900db2878e8ff2b85d0a65b813be4833d0", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "97a36d8770ae4ef4f5687d602a5b3e863260ebc570172a3067f12c51120a413c", "generation_seconds_batch": 23.56437352800276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.964149236679077, "target_z": 1.781250031147121, "ce": 6.006728172302246, "genuineness": 3.747946262359619, "identity": 0.5315150022506714, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/736d7ee32341a3c43f0d0ed6f92e4c900db2878e8ff2b85d0a65b813be4833d0.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "737e0f29a90fa164cd40e68f87c56868722466a2a7668d478b9bc7a5515079c3", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "2ee16daa79d59ecdbd5f2c5cf9a242938ca94dfe2997aba67eff83d494a2e3d1", "generation_seconds_batch": 28.764786492916755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.098702430725098, "target_z": 1.8749998754115158, "ce": 6.252951145172119, "genuineness": 3.6412534713745117, "identity": 0.20240303874015808, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/737e0f29a90fa164cd40e68f87c56868722466a2a7668d478b9bc7a5515079c3.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "7412666bf8294ba99b8ac188e6e202fd13a86e24e068cfb9aa38e59cf2ef96ee", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "c0a1e9c550140165ca9fc3385953e5f3bf99c62ba0a092b542278b36671527d1", "generation_seconds_batch": 22.345885685062967, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9305107593536377, "target_z": 1.7578124454925381, "ce": 6.38003396987915, "genuineness": 4.331617832183838, "identity": 0.31886109709739685, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7412666bf8294ba99b8ac188e6e202fd13a86e24e068cfb9aa38e59cf2ef96ee.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "7413e1f8ff278b54f062de3224f4d5a9f0767f7134d9dcffeeddd0768dc281a9", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "1bcdc7a4f5ccf25937cccaa2573be64b7448083ffba2086886e45b74eb4afaaa", "generation_seconds_batch": 24.308317846036516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.313807725906372, "target_z": 1.328125036338308, "ce": 6.09409761428833, "genuineness": 3.848362922668457, "identity": 0.31277167797088623, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7413e1f8ff278b54f062de3224f4d5a9f0767f7134d9dcffeeddd0768dc281a9.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "741726330450390ab379839e77fb98bf5451d35fb978308b99e39daaee071d66", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "f2431be42ea729a11982280fe2486853556989e896d9d7a75107798930598eb2", "generation_seconds_batch": 23.870866591925733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.336233139038086, "target_z": 1.3437499273233842, "ce": 5.9431867599487305, "genuineness": 3.9425034523010254, "identity": 0.6997559666633606, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/741726330450390ab379839e77fb98bf5451d35fb978308b99e39daaee071d66.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "744439a36bc0675d883a67a73b51a83f4a1052bcb3fcd5d239652c3fd544a7f6", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "b3a2d26a6ec43fc2e4822c89fea3fb3cb8f2e7a710e3f97c16212135fa3a55a3", "generation_seconds_batch": 24.205523965065368, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.221180438995361, "genuineness": 3.7667744159698486, "identity": -0.011174088343977928, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/744439a36bc0675d883a67a73b51a83f4a1052bcb3fcd5d239652c3fd544a7f6.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": "74b57416958ecac3cdee4a8b2e11368cf2fd38842e9bf65c90dac5aaba622def", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f88054b887340447d959123ef965de26beb48e82f6359413cdd284345c7ed370", "generation_seconds_batch": 25.09835938096512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2108306884765625, "target_z": 1.9531251609267921, "ce": 6.356205940246582, "genuineness": 3.5973212718963623, "identity": 0.36163705587387085, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/74b57416958ecac3cdee4a8b2e11368cf2fd38842e9bf65c90dac5aaba622def.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "74c16b3cd2612ea60de695423bd74465ddc3d382ac8b43854f0197f993b241df", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "20b04e43ef39cf2f8603f0925d3c87c6c4cf74291ec6a399119b5b2370b6e539", "generation_seconds_batch": 24.352012494928204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7847447395324707, "target_z": 1.6562500726766158, "ce": 6.205295085906982, "genuineness": 4.344170093536377, "identity": 0.5568060874938965, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/74c16b3cd2612ea60de695423bd74465ddc3d382ac8b43854f0197f993b241df.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "74e99fd87ee5cb743ab4d9e9fffaddc2353b06315ffbaefa12cc28132753983d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3becc53fa46a8dd4e98a7f1860f9141ef4e3b6403bf18b397656c98b80a962a0", "generation_seconds_batch": 23.069037634995766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.504425048828125, "target_z": 1.4609375233603408, "ce": 5.990842819213867, "genuineness": 4.030367851257324, "identity": 0.3687615692615509, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/74e99fd87ee5cb743ab4d9e9fffaddc2353b06315ffbaefa12cc28132753983d.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "75a33d90fd9e4680ba52c0c1edd6a60a7dbec254ffff8f239d12b081b54e7a1f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "127fcd9e700ddd57433d0f54890a1f1eed3dee0bea1efa22041a35cecdd19884", "generation_seconds_batch": 25.53700414195191, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.636916160583496, "target_z": 2.2499999169410105, "ce": 6.173524379730225, "genuineness": 4.262581825256348, "identity": 0.6832054257392883, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/75a33d90fd9e4680ba52c0c1edd6a60a7dbec254ffff8f239d12b081b54e7a1f.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": "7671dd76646671f37d8e928bcc667f30e14e4acafdc46f748c10122204f4f47f", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I wrote the number on the envelope.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a650783b8fdb9a64097ed29f61927eabcbc9b14a6cd7bbc7e50d3cb5f393cda7", "generation_seconds_batch": 22.267951305024326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.627765655517578, "target_z": 1.5468750051911868, "ce": 6.324435234069824, "genuineness": 3.9048471450805664, "identity": 0.21168668568134308, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7671dd76646671f37d8e928bcc667f30e14e4acafdc46f748c10122204f4f47f.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "7745a65c63c3a05be280a81d0c450b451acadecfc3be31978bf1ac6332607068", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "af94ff787c6a3549e8466bacc7e68a75dfbb22cef6f3547d07394cc3d198a53b", "generation_seconds_batch": 23.258987743989564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.908085346221924, "target_z": 1.7421875545074619, "ce": 6.268836498260498, "genuineness": 4.093128204345703, "identity": 0.4958655834197998, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7745a65c63c3a05be280a81d0c450b451acadecfc3be31978bf1ac6332607068.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "774b36dabd4121930920a16f4c6b2aab46ffdca4f1bd8ceb51f189a874583eab", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "da8e996dd5ee00139fba0e6d9a18085de7fdd190e4acc1191c3cc603096dc67c", "generation_seconds_batch": 28.764786492916755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 6.23706579208374, "genuineness": 3.528284788131714, "identity": 0.26807063817977905, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/774b36dabd4121930920a16f4c6b2aab46ffdca4f1bd8ceb51f189a874583eab.mp3" }, { "id": "heldout-fr-3", "language": "fr", "sentences": [ "Quelqu’un a déplacé la petite chaise bleue.", "Elle était ici... près de la grande bibliothèque.", "Nous la remettrons après le déjeuner." ], "text": "Quelqu’un a déplacé la petite chaise bleue. Elle était ici... près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-3", "key": "77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c543b16b425333b735cd34196f8846b759d138e12fde47d2ac391fbe84f2e589", "generation_seconds_batch": 17.881119925063103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.547214031219482, "target_z": 2.1875000207647473, "ce": 6.125868320465088, "genuineness": 4.362998008728027, "identity": 0.8360499143600464, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "77c1a9fceeb652f3fc318a229a39a709d9df8addeda94395ba0f71a4e1d43ed7", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "ce828f14532ea5d4cf2c733eb7ebf20e399fa6b8248d18a5571b5f0d6603cd83", "generation_seconds_batch": 15.755182291031815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6165528297424316, "target_z": 1.5390624766396592, "ce": 6.213237762451172, "genuineness": 4.369274139404297, "identity": 0.5106967687606812, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/77c1a9fceeb652f3fc318a229a39a709d9df8addeda94395ba0f71a4e1d43ed7.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": "77e5c83a31f252a228843d0b1994f00d3897c8aba3759d5de4211d4f7f3ff660", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9a5803ea3d51dff2ed65f23aaa0eeb0d87256eeff8a680a0b4088526603fa5a1", "generation_seconds_batch": 18.15453819592949, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 6.125868320465088, "genuineness": 4.268857479095459, "identity": 0.65678870677948, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/77e5c83a31f252a228843d0b1994f00d3897c8aba3759d5de4211d4f7f3ff660.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "77f1432e668afca80b838989d7ae5cab67d0f5e63a5461e8ffe1872a9b1815ce", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "a830c98ee69048895e156eb5eaa80792e765f8c28ab6a6d2c358f2fe0ae072c5", "generation_seconds_batch": 18.027226150035858, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.313807725906372, "target_z": 1.328125036338308, "ce": 6.173524379730225, "genuineness": 4.18099308013916, "identity": 0.4902275800704956, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/77f1432e668afca80b838989d7ae5cab67d0f5e63a5461e8ffe1872a9b1815ce.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": "77f3a1cc9957b4aa63278c24ce2ac38b12c7c7bb8316e912b20efd83d524bb4c", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c6c6d9ef8851972ef010fc81a970446ec7e95570357603e43c1f22a13fd09a9", "generation_seconds_batch": 26.32801786600612, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.614490509033203, "target_z": 2.2343748598379554, "ce": 6.0305562019348145, "genuineness": 4.005263805389404, "identity": 0.7455622553825378, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/77f3a1cc9957b4aa63278c24ce2ac38b12c7c7bb8316e912b20efd83d524bb4c.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": "78473bbe9796043f0f9e34f05dc7985af2230c4e9e6b4aab4d53ea47db00f6b4", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "f053454a52e2d2b08fd519ba89640f75d407238ace76117dbaf4ac08bb87e3d4", "generation_seconds_batch": 18.25528566900175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.086154937744141, "genuineness": 4.137060642242432, "identity": 0.5683514475822449, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/78473bbe9796043f0f9e34f05dc7985af2230c4e9e6b4aab4d53ea47db00f6b4.mp3" }, { "id": "heldout-nl-1", "language": "nl", "sentences": [ "De sleutels liggen op de keukentafel.", "Ik keek in de la... en daarna naast het raam.", "Neem ze mee als je weggaat." ], "text": "De sleutels liggen op de keukentafel. Ik keek in de la... en daarna naast het raam. Neem ze mee als je weggaat.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-1", "key": "794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2085c4c966390b576a117070b8075940c009ce38758080accf9ae9378c475ffd", "generation_seconds_batch": 20.45129947399255, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.076276779174805, "target_z": 1.8593748183084604, "ce": 6.0305562019348145, "genuineness": 4.20609712600708, "identity": 0.8104366064071655, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "79a8bc92ab4e675f4efa0a97f78e126880821eb313156d5d4718d9c8f5cd129b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7de02564227e3b89daa776322bcf9fac13c75e6d950e039fdc214d56fb0700c2", "generation_seconds_batch": 16.348945526056923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5604889392852783, "target_z": 1.5, "ce": 5.871703147888184, "genuineness": 3.9801597595214844, "identity": 0.4765818119049072, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/79a8bc92ab4e675f4efa0a97f78e126880821eb313156d5d4718d9c8f5cd129b.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": "79ebc676fca199b5fe6a0ba6a649f0ae0bd480c0046349a4312c66d4e9494161", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I wrote the number on the envelope.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "852a5b99bf64caf61b443b3faed61850966c9e5f49408c9026e9c9917560dbdf", "generation_seconds_batch": 22.508398242061958, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0202131271362305, "target_z": 1.8203125077867803, "ce": 6.252951145172119, "genuineness": 3.6224255561828613, "identity": 0.36051344871520996, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/79ebc676fca199b5fe6a0ba6a649f0ae0bd480c0046349a4312c66d4e9494161.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "7a1fe7c7112a8bcb1e5c64ec816ef0d3ec64f24f750361fa36aa8f88029c6685", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f3e5890e1ae94604062801c944d6e0974ddcb535406f7aa73fc2628cf342e49b", "generation_seconds_batch": 24.31624027399812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8744468688964844, "target_z": 1.718749968852879, "ce": 5.863759994506836, "genuineness": 4.3127899169921875, "identity": 0.6015282869338989, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana, vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7a1fe7c7112a8bcb1e5c64ec816ef0d3ec64f24f750361fa36aa8f88029c6685.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "7a65307dd3a82c02c284eada9130e04f8a5a705b015e202878c59e4fcac3df71", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "3e366ecaf83f61886713e856b64239f61e36396c1ba58f9e5e107412cbf76e36", "generation_seconds_batch": 24.18446216103621, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1119775772094727, "target_z": 1.1875000207647475, "ce": 6.0305562019348145, "genuineness": 4.042920112609863, "identity": 0.315369576215744, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7a65307dd3a82c02c284eada9130e04f8a5a705b015e202878c59e4fcac3df71.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "7b4e92b04e735196f862bfb2799ceef8535bb40a69fe4a25364dc2a4498f7e43", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "17b779f2d6983a0d034a5e4f157b9471c8ca8ecce70f0866744db0474be742c6", "generation_seconds_batch": 23.917659475002438, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0559134483337402, "target_z": 1.148437378007109, "ce": 6.300607204437256, "genuineness": 4.086852550506592, "identity": 0.6649580597877502, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7b4e92b04e735196f862bfb2799ceef8535bb40a69fe4a25364dc2a4498f7e43.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "7b650a4a10f0487f10b74fcd4116686a4c9f19a7a4a9ebd993ed6aed200232ca", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "98b5a119658cc5504135b9493aba2f1a0905c9a5a8245ace9d266c277c0ab2ba", "generation_seconds_batch": 27.103716801968403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25568151473999, "target_z": 1.9843749428969448, "ce": 6.086154937744141, "genuineness": 4.049196243286133, "identity": 0.6751655340194702, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7b650a4a10f0487f10b74fcd4116686a4c9f19a7a4a9ebd993ed6aed200232ca.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "7bdfcda9ef41f25567d6043cbffb76d85a4b0f6e9874482a7fef2b4998b3e52c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f11eee5d7ba75be11271819493412ca087b2c0ab8d351f3360234d7a31f486ca", "generation_seconds_batch": 23.928455842891708, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.076276779174805, "target_z": 1.8593748183084604, "ce": 6.117925643920898, "genuineness": 3.9927117824554443, "identity": 0.24528393149375916, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7bdfcda9ef41f25567d6043cbffb76d85a4b0f6e9874482a7fef2b4998b3e52c.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "7c3073011be357f3ba2abb0ac3f2416cb3f6dd684be568b3380fa8b4d9cdbd20", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "257295554a05d0e31485c9a152b27af51ef97065564fe09dce229c6b2d516850", "generation_seconds_batch": 20.3937280169921, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.076276779174805, "target_z": 1.8593748183084604, "ce": 6.1020402908325195, "genuineness": 4.099404335021973, "identity": 0.8160939812660217, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7c3073011be357f3ba2abb0ac3f2416cb3f6dd684be568b3380fa8b4d9cdbd20.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "7c3af0e07aa47bb1711881be13ecb4e897a606897578c79eec858973c4931eba", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d57520e5fe874f13ab45d9deb58273ddc6b0f56461936e0a88de93350714d2ca", "generation_seconds_batch": 24.258738000062294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.177192211151123, "target_z": 1.9296875752722091, "ce": 6.252951145172119, "genuineness": 4.406930446624756, "identity": 0.6505447626113892, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7c3af0e07aa47bb1711881be13ecb4e897a606897578c79eec858973c4931eba.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "7c8826b87b5872c2247500d67e3ee8df33ea6205d82b744535d7ce2508aedf52", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "5bc32c36b924b3570dd9375dc66c1ab9f7d0e4ff14f6b41b1d785e5042ac2ddb", "generation_seconds_batch": 24.27685255196411, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6838295459747314, "target_z": 1.585937481830846, "ce": 6.165581703186035, "genuineness": 4.306513786315918, "identity": 0.6125074625015259, "hypothesis": "De deur staat nog open. Ik heb je Jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7c8826b87b5872c2247500d67e3ee8df33ea6205d82b744535d7ce2508aedf52.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "7c9de2c7a15a556a32a7ea6755b9edc7ae1e10080b5e7614d1c1588d86fc80b7", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "86bc91834185513b458d96370de9355e948f91f474559a5a4e3a27c8304b9380", "generation_seconds_batch": 23.143791874987073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7398934364318848, "target_z": 1.6249999584705053, "ce": 6.109982967376709, "genuineness": 4.086852550506592, "identity": 0.7177336812019348, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7c9de2c7a15a556a32a7ea6755b9edc7ae1e10080b5e7614d1c1588d86fc80b7.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "7cac8c69c9afa23087e51f4d459faf29b18d7f6bd4bd1452a10e08fcf32e26d1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "1cd20509e55a34c46c72447d5e2b8d5835c92241858aa116934bd81bd78b0581", "generation_seconds_batch": 24.515246671973728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2108306884765625, "target_z": 1.9531251609267921, "ce": 6.070269584655762, "genuineness": 4.218649387359619, "identity": 0.5822594165802002, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7cac8c69c9afa23087e51f4d459faf29b18d7f6bd4bd1452a10e08fcf32e26d1.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "7cd32d22930e4e2650bf7630226a0ed93598000a7ecf1e04e156291186b8189c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "8d1a596a15c7fdcd135aeffafff125d0f7cfd63e5b8eec7ba3b6bc41d8e8304a", "generation_seconds_batch": 24.535420494969003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.502363204956055, "target_z": 2.1562502387945948, "ce": 6.221180438995361, "genuineness": 4.011539936065674, "identity": 0.3461247682571411, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7cd32d22930e4e2650bf7630226a0ed93598000a7ecf1e04e156291186b8189c.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "7cdd1ba12569b2489acbf516da1799df591d87562e27e7d2dd80143621080041", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4e4677cf2f4a2d4745b4eaed9300df98ef8712afc6821f86a94cce9a0a697427", "generation_seconds_batch": 16.979927664040588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8295958042144775, "target_z": 1.6875000207647475, "ce": 5.71682071685791, "genuineness": 3.9927117824554443, "identity": 0.485308974981308, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7cdd1ba12569b2489acbf516da1799df591d87562e27e7d2dd80143621080041.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "7d048fba2fbabf317bfcff148c88e9d6702ea205dadadf135b1cba31b7bd848c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "290c181915370e2eeb72252cbcbc679f5b0a3cac1b655f30e80f27740c458147", "generation_seconds_batch": 24.270309402025305, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.605340003967285, "target_z": 1.5312499480881316, "ce": 6.173524379730225, "genuineness": 4.055472373962402, "identity": 0.5761405229568481, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7d048fba2fbabf317bfcff148c88e9d6702ea205dadadf135b1cba31b7bd848c.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": "7d49aea3befb7df87439b2d8d7d637b4b7e17d877fc8650b14e49d13c087074d", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The road turns left behind the bridge.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "915c6be661d38d0ce6597b66d6ee890e41aca69bc994041e430558a851b4da94", "generation_seconds_batch": 20.27465147699695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.695042371749878, "target_z": 1.5937500103823736, "ce": 6.356205940246582, "genuineness": 3.8169827461242676, "identity": 0.3112044930458069, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7d49aea3befb7df87439b2d8d7d637b4b7e17d877fc8650b14e49d13c087074d.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "7d64b38c504070e0641a5d8ccfe063a50a80713c23145f1996a589572e8e4fca", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "91ad1857915f2d24b2f178f3057a69466a4168aebbe4a4c2457f2c48b0503b2a", "generation_seconds_batch": 27.131078917067498, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5829145908355713, "target_z": 1.5156250571030552, "ce": 6.395919322967529, "genuineness": 4.080576419830322, "identity": 0.7561836838722229, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7d64b38c504070e0641a5d8ccfe063a50a80713c23145f1996a589572e8e4fca.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": "7db2fd9c42baf25c5c88c8f47e129a705ac187ca78ad19539ae13e3be2dd60dc", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cb833b08424dd4b97c7995c30cbca31efb2c1cc194544ae6232f794276f8bf58", "generation_seconds_batch": 23.353118296014145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.372091293334961, "genuineness": 4.306513786315918, "identity": 0.46689945459365845, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7db2fd9c42baf25c5c88c8f47e129a705ac187ca78ad19539ae13e3be2dd60dc.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "7dba8598df0a8770d1c07da15c90f73adaba684232e468f0c018915a395c763f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "64df45e0bf90618c0b9bdce119d2a693ba1b9e80e899feef24857773c72608f3", "generation_seconds_batch": 21.860672302893363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.795957326889038, "target_z": 1.6640624351101645, "ce": 5.982900142669678, "genuineness": 4.118232727050781, "identity": 0.6208378076553345, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7dba8598df0a8770d1c07da15c90f73adaba684232e468f0c018915a395c763f.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": "7ddc9d357be2aee0f7f71f34fc1c0188781b2b6e347092e6befa8d0c72062d69", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Escribí el número en el sobre.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6465f8d0e0010ac561fc59df46cf3115feaf66a914a9be7f6bc118374a029dda", "generation_seconds_batch": 22.592100306996144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.053851127624512, "target_z": 1.8437497612054052, "ce": 6.046441555023193, "genuineness": 4.118232727050781, "identity": 0.7768423557281494, "hypothesis": "Escribí el número en el sobre, léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7ddc9d357be2aee0f7f71f34fc1c0188781b2b6e347092e6befa8d0c72062d69.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "7de3e3c62b427524227aeb36f7ad4f20565f9bd8843a2ef66af853844a3666e5", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "a92e9cea790992c2a7502b7b23b1a27130346d99f5cc1bacf330ab5cdaf9d406", "generation_seconds_batch": 17.12073755392339, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.042638778686523, "target_z": 1.8359375648898355, "ce": 6.443574905395508, "genuineness": 4.519899368286133, "identity": 0.41555845737457275, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7de3e3c62b427524227aeb36f7ad4f20565f9bd8843a2ef66af853844a3666e5.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": "7e336afaadad9a59850c7b87b83ec40b7c2dc01f5a935628668be39a9a2aa27e", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Escribí el número en el sobre.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7654ae37f7c6bede5c4b8e79691fb199171109e745474217299df1c3a6aae68c", "generation_seconds_batch": 22.592100306996144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8968725204467773, "target_z": 1.7343750259559343, "ce": 6.229123115539551, "genuineness": 4.344170093536377, "identity": 0.7066920399665833, "hypothesis": "Escribí el número en el sobre. Léelo bien. Y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7e336afaadad9a59850c7b87b83ec40b7c2dc01f5a935628668be39a9a2aa27e.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "7e4e7e91f723ffaa25bdbfa9dd633ca69df5d0f3176ff475031fedb4a869e637", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "91f9323212dbc6f3f20d291d1086724c360952ff119186b72c20cd47f6c8ce7d", "generation_seconds_batch": 23.091814556973986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.437148332595825, "target_z": 1.414062518169154, "ce": 5.935244083404541, "genuineness": 3.873466968536377, "identity": 0.677298903465271, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7e4e7e91f723ffaa25bdbfa9dd633ca69df5d0f3176ff475031fedb4a869e637.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "7eb45c3959473a586996d2c39cac218e195fbfff6bd4f60c9be369834ec499e1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "3661d61bf38c6effaf3812c87cb48d92e768624ed23761c1d9bc962c23291d56", "generation_seconds_batch": 26.718650197959505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8295958042144775, "target_z": 1.6875000207647475, "ce": 6.109982967376709, "genuineness": 4.143336772918701, "identity": 0.4971729516983032, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7eb45c3959473a586996d2c39cac218e195fbfff6bd4f60c9be369834ec499e1.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "7ec3e82414ed716771c0e67572f1dea809b85181f97f040415f124da00447e63", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "7a73f2987be9ff8c6d164f58965893390ad802c44a0ac013c283efa8f1c3afc8", "generation_seconds_batch": 24.593137993942946, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.031425952911377, "target_z": 1.828125036338308, "ce": 6.252951145172119, "genuineness": 4.2249250411987305, "identity": 0.273661345243454, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7ec3e82414ed716771c0e67572f1dea809b85181f97f040415f124da00447e63.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9a8a8756ec50943307fb35e6a9d6d35a7c3b8a565d55b7d2a0e9142bd79d9d8f", "generation_seconds_batch": 18.23727683199104, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.291382074356079, "target_z": 1.3124999792352525, "ce": 6.3641486167907715, "genuineness": 4.457139015197754, "identity": 0.40348175168037415, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239.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": "7fcacd0df5704a29d1e09c4febe2beef4588e6c8aafe08d6e80961deb0385bc4", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "39f290ab4e992b812734c639a677fe99fed6f0fa843470f9d6a794919992feaa", "generation_seconds_batch": 18.12534519098699, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.300532817840576, "target_z": 2.015625057103055, "ce": 6.475345611572266, "genuineness": 4.356721878051758, "identity": 0.50742107629776, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7fcacd0df5704a29d1e09c4febe2beef4588e6c8aafe08d6e80961deb0385bc4.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "7fe5943f680067fd85f1379952b74a1ba8e086f178267de859cbac9a16e952d5", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "f0b00af83eb462a747ae6a2a5035bfffd4caee75c47bc7ea758fb14bd2a88410", "generation_seconds_batch": 24.221068004961126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.278107166290283, "target_z": 2.0, "ce": 6.5786004066467285, "genuineness": 4.306513786315918, "identity": 0.6093628406524658, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/7fe5943f680067fd85f1379952b74a1ba8e086f178267de859cbac9a16e952d5.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "806111e30f3a84e61d1ab632c5e6eb1dbfe041101b7b964f31f68a36d291d8c9", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "1986302b7d4f1257435dbb46278bbb560733d4a2693aeda582668901aaeeddb8", "generation_seconds_batch": 23.548058814019896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.627765655517578, "target_z": 1.5468750051911868, "ce": 6.09409761428833, "genuineness": 3.625563621520996, "identity": 0.26370012760162354, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/806111e30f3a84e61d1ab632c5e6eb1dbfe041101b7b964f31f68a36d291d8c9.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "371494fae22609733c378778361dcc3a807f48449705045e3f42ddcae91c0811", "generation_seconds_batch": 17.87595609703567, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5604889392852783, "target_z": 1.5, "ce": 6.24500846862793, "genuineness": 4.457139015197754, "identity": 0.6163174510002136, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "80d232751cfd63fc3320657cf507657e5e89b37c1290cc5d9dd487278c401ac1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "deec431e2c0306807763685920185b5e38d822a4095cca95284a03e54824fed6", "generation_seconds_batch": 24.58658734499477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7623190879821777, "target_z": 1.6406250155735604, "ce": 6.324435234069824, "genuineness": 4.281409740447998, "identity": 0.570458173751831, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/80d232751cfd63fc3320657cf507657e5e89b37c1290cc5d9dd487278c401ac1.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "813c5bad1115d30795bd03f9ee6321c0ba59bf3e3747276a0f6552da1de22e8e", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "483058b1be706e434abbf9c81154680b6166aea95f1aff7f51724168897a32e6", "generation_seconds_batch": 23.769994407892227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.695042371749878, "target_z": 1.5937500103823736, "ce": 6.197352409362793, "genuineness": 4.20609712600708, "identity": 0.31561875343322754, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/813c5bad1115d30795bd03f9ee6321c0ba59bf3e3747276a0f6552da1de22e8e.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "81675f48f042578e61ccd8d14204ddb43886ed5097119b3064cf31ceaf2760cf", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2a9e49354944e28412af1735efedded8808dbe59c343022f22ee719cd44323d0", "generation_seconds_batch": 20.754844077979214, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1568286418914795, "target_z": 1.218749968852879, "ce": 5.839932441711426, "genuineness": 3.8922953605651855, "identity": 0.6189627051353455, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/81675f48f042578e61ccd8d14204ddb43886ed5097119b3064cf31ceaf2760cf.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "81b8868bea0d7babe7b4d2c6c833e6d0f448fae9d6dde27c77d6126359f960cf", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "eee14867c5963a7dda2c1747488a0899c440d3dd6beb1c90246153947dec53ea", "generation_seconds_batch": 25.659792028018273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.504425048828125, "target_z": 1.4609375233603408, "ce": 6.205295085906982, "genuineness": 4.231201171875, "identity": 0.5081009864807129, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/81b8868bea0d7babe7b4d2c6c833e6d0f448fae9d6dde27c77d6126359f960cf.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "81fb1ce33e5092417818b8ea7d88401bf80e23cb695211f2fdacd913ffb69e87", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "843b9cdfa32b5abe85d094071adc749dd80c28c11b1fd6a0af3da34f36c7e5d9", "generation_seconds_batch": 24.67765392397996, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9192981719970703, "target_z": 1.7500000830589895, "ce": 5.8875885009765625, "genuineness": 4.068024158477783, "identity": 0.46482980251312256, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/81fb1ce33e5092417818b8ea7d88401bf80e23cb695211f2fdacd913ffb69e87.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "82de1ff1848e05e7a29fd153facd9dbd1ec9ff3eb67de2fab80ccb32495911ef", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "15c218b460918d805a8edecf687c2325c7ecaef551a77fccd4eeae8fd45d1be8", "generation_seconds_batch": 21.6687438840745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.479937553405762, "target_z": 2.1406251816915396, "ce": 6.435632705688477, "genuineness": 4.118232727050781, "identity": 0.5956577062606812, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/82de1ff1848e05e7a29fd153facd9dbd1ec9ff3eb67de2fab80ccb32495911ef.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "82f14ba8f1454e69f235a6c8c6d18accd9a871558cd735b9230a88e3bf15b7b1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c079df38c1e755250b1edba8603a542eb170cc1bc44edaed48d522e3c4dcedc3", "generation_seconds_batch": 25.509166829986498, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.502363204956055, "target_z": 2.1562502387945948, "ce": 6.348263263702393, "genuineness": 4.168440818786621, "identity": 0.7874215841293335, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/82f14ba8f1454e69f235a6c8c6d18accd9a871558cd735b9230a88e3bf15b7b1.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "83402e8138877b42595a40618743183c73b72c7dcfdcf291a0953df4e00a81c5", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "0bc3e77711ac1f1fc088962e170ff565380f1437f0ddd444a4b4c2389bb36f72", "generation_seconds_batch": 21.411808006931096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.044700860977173, "target_z": 1.1406250155735604, "ce": 5.990842819213867, "genuineness": 4.13078498840332, "identity": 0.09226561337709427, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/83402e8138877b42595a40618743183c73b72c7dcfdcf291a0953df4e00a81c5.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "834addcfdf4d5ba4f6fead6cff56c9d8ef5b393df6e107464edc00d3ca959fba", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "9bbdf24329cd2a68eaa416dcf19e53104da3258a0d61087c13651d075e01565d", "generation_seconds_batch": 23.03402772604022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0895519256591797, "target_z": 1.171874963661692, "ce": 5.871703147888184, "genuineness": 3.747946262359619, "identity": 0.37284135818481445, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/834addcfdf4d5ba4f6fead6cff56c9d8ef5b393df6e107464edc00d3ca959fba.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "834b53930d01f8d92af157d13ab9e11ea4a245dd28d791476a9cfc358f92aa56", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "09160db54b425628aec9e41acf93c6f619214768645c4a00202b0c42dd0c7fa5", "generation_seconds_batch": 18.055538668995723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7623190879821777, "target_z": 1.6406250155735604, "ce": 6.173524379730225, "genuineness": 4.331617832183838, "identity": 0.4649914801120758, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/834b53930d01f8d92af157d13ab9e11ea4a245dd28d791476a9cfc358f92aa56.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "8359724ac2b6e3cd5f14e5b44ddf457392b8acb7465ab55cad3687081c7db4ea", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "87079e00acd7ac9e34243bbc893d07a3b274959ac06c3ca85702d5de3d739a25", "generation_seconds_batch": 20.127385213971138, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 6.205295085906982, "genuineness": 4.193545341491699, "identity": 0.7546759843826294, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/8359724ac2b6e3cd5f14e5b44ddf457392b8acb7465ab55cad3687081c7db4ea.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "836fcd516d94e58645bdb057c14c967fa7d264d58b3c6c96fbe1567f8e158459", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4ef7ee13de864eb8f9f5f1cd1ceaa43eb720a83cebd08d00973313bebef748ba", "generation_seconds_batch": 23.548058814019896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7847447395324707, "target_z": 1.6562500726766158, "ce": 6.054384231567383, "genuineness": 3.8546390533447266, "identity": 0.17999228835105896, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/836fcd516d94e58645bdb057c14c967fa7d264d58b3c6c96fbe1567f8e158459.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "83fcebb80f9a560dae4dd24d5efa9239ffd32a77f755da75270a6e464196f53f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "acb5b6b576778d56ffdb31af2119239a0cdc48c6622c893abc411b1b498cfcff", "generation_seconds_batch": 24.512330768979155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.840808629989624, "target_z": 1.695312549316275, "ce": 6.260893821716309, "genuineness": 4.394378662109375, "identity": 0.4724442958831787, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/83fcebb80f9a560dae4dd24d5efa9239ffd32a77f755da75270a6e464196f53f.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": "842cd407c07066188554b6519d906833e2133578ac023717e9e2b3efa9c6694b", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Your message arrived after I had left.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8520c6a700589d2fe8eb6a7959b0e11ccd6270df8356e4008e689305f0ae9311", "generation_seconds_batch": 24.946528493077494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2016797065734863, "target_z": 1.2499999169410105, "ce": 6.419747352600098, "genuineness": 3.873466968536377, "identity": 0.34930548071861267, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/842cd407c07066188554b6519d906833e2133578ac023717e9e2b3efa9c6694b.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "843e3b3f1337958545b21956ad95b4877d4ea4270a0b6e0284c546d5aba84ac5", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "1621e4b1e91b783b5d9c570cc95ea5137be710b55e1ccb16a07665fab9e1893e", "generation_seconds_batch": 23.539477342972532, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.717468023300171, "target_z": 1.609375067485429, "ce": 5.95112943649292, "genuineness": 3.823258876800537, "identity": 0.3824703097343445, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/843e3b3f1337958545b21956ad95b4877d4ea4270a0b6e0284c546d5aba84ac5.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "858ea148739c577b6c9269f88899e9ddd7c619d0c5243cddff5a858c0d2cd661", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f8b6ec3861185a1121dffcad8a796453b3c1706f83d8fa08a3d47f51f098b37e", "generation_seconds_batch": 24.38150791800581, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.109915733337402, "target_z": 1.8828127361990015, "ce": 6.300607204437256, "genuineness": 3.7542223930358887, "identity": 0.33033788204193115, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/858ea148739c577b6c9269f88899e9ddd7c619d0c5243cddff5a858c0d2cd661.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": "85a755bc6367f3df9cbf56dc0cde242d542fd93ce9b90808bd35c81c2441ec13", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "134df1de6d2ea68feae5b40d79dcba39fca34e721575fa2ec487ef2cb2795e26", "generation_seconds_batch": 23.256196181988344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.695042371749878, "target_z": 1.5937500103823736, "ce": 6.308549880981445, "genuineness": 4.193545341491699, "identity": 0.4688279628753662, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg, sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/85a755bc6367f3df9cbf56dc0cde242d542fd93ce9b90808bd35c81c2441ec13.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "85c839fada1016e48c1c5a4fca4416a335ff0879f6f879be4679bcbe9357df71", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "ef19999de54d234b9168ce2473e1a3dcb579563e475c8e71c062ac894faa2388", "generation_seconds_batch": 17.151270417962223, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7847447395324707, "target_z": 1.6562500726766158, "ce": 6.23706579208374, "genuineness": 4.419482707977295, "identity": 0.652583122253418, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/85c839fada1016e48c1c5a4fca4416a335ff0879f6f879be4679bcbe9357df71.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "85d73b55454fe42c8d56cff6465a18acae04816e5def1d11ab4d341cfb5dedb5", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "279616454785381e8e1906266bc791b284bc1e50e168d7361b2ab72cc21cccb8", "generation_seconds_batch": 24.565812389017083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.358658790588379, "target_z": 1.3593749844264396, "ce": 6.221180438995361, "genuineness": 3.986435890197754, "identity": 0.43091878294944763, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/85d73b55454fe42c8d56cff6465a18acae04816e5def1d11ab4d341cfb5dedb5.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "86346807c75c2616f985a92c6d1ef69fd1f2ca7a89d801f8687354fc195e3c0a", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "c1305bf344e44b8f2d24e08ed133c433b346b6847c96ae64ca4e58d15a39bf78", "generation_seconds_batch": 29.00909856299404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165979385375977, "target_z": 1.9218750467206815, "ce": 5.808161735534668, "genuineness": 3.8922953605651855, "identity": 0.3619081974029541, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/86346807c75c2616f985a92c6d1ef69fd1f2ca7a89d801f8687354fc195e3c0a.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "86e7518abcc23a2a08aa79d0b938902d53f7fad7faa70476c2967f9c3bba2061", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "e11297a3eaed3dd3c599b55ec1a6cac0d00eb105f83895ebd9c6ba69fbe1ea53", "generation_seconds_batch": 26.496915864991024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.479937553405762, "target_z": 2.1406251816915396, "ce": 6.038498878479004, "genuineness": 3.8671910762786865, "identity": 0.42757415771484375, "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_S_NARR_reg/86e7518abcc23a2a08aa79d0b938902d53f7fad7faa70476c2967f9c3bba2061.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "86ff109d779b8003663dffacdcef92c984c1c07e1ceba9ce6d01a5fc57c29433", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "8189d4f146a198162295eb7619867d2c749042547e1ccb11a1fa45196d9e7dd3", "generation_seconds_batch": 23.471925292047672, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7623190879821777, "target_z": 1.6406250155735604, "ce": 6.0305562019348145, "genuineness": 4.168440818786621, "identity": 0.7251216769218445, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/86ff109d779b8003663dffacdcef92c984c1c07e1ceba9ce6d01a5fc57c29433.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "871716f0f220fa12464648220f3a54b84d74328bf5d8aefe099a0d7215a360b0", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "a192298dd5b708b24442c0e8770fe05a743787512b7ea801a879e464ef75f2e7", "generation_seconds_batch": 23.342685795971192, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.479937553405762, "target_z": 2.1406251816915396, "ce": 6.09409761428833, "genuineness": 4.068024158477783, "identity": 0.7419146299362183, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/871716f0f220fa12464648220f3a54b84d74328bf5d8aefe099a0d7215a360b0.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "8803bd8719247e53c8bc5d8cdbb91df48de14af2be6852a2c24ad8838e7010bf", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "a48b2b191edd48badca9cec87bb198cc1b250afc312d889b6011319324275f27", "generation_seconds_batch": 18.10032077797223, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.547214031219482, "target_z": 2.1875000207647473, "ce": 6.316492557525635, "genuineness": 3.773050308227539, "identity": 0.3715994954109192, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/8803bd8719247e53c8bc5d8cdbb91df48de14af2be6852a2c24ad8838e7010bf.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": "880aee3b1fce628b43567252f0399e01df5d88e7da0711bb93b4cfa9ad8c8e76", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "321aa2f157a424744941f038150e55897873747f15eae7998c58c8d873a088ae", "generation_seconds_batch": 18.76157329895068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.885659694671631, "target_z": 1.7265624974044065, "ce": 6.09409761428833, "genuineness": 4.080576419830322, "identity": 0.4890456199645996, "hypothesis": "De weg gaat linksaf na de brug. Loop verder. Tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/880aee3b1fce628b43567252f0399e01df5d88e7da0711bb93b4cfa9ad8c8e76.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "8880b51ae97963655df886e716eb3793872b3eff4a7bbb46dcc23c1494116385", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "b58162e3f74051f0346b2d164aac5aedc8b1eb9711167c979683209431600fd0", "generation_seconds_batch": 24.007206290960312, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.502363204956055, "target_z": 2.1562502387945948, "ce": 6.165581703186035, "genuineness": 4.482243061065674, "identity": 0.5371988415718079, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/8880b51ae97963655df886e716eb3793872b3eff4a7bbb46dcc23c1494116385.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "889138f80cb5b5041e0e425162f0c434a68f9fd7ef9574622646b98e93b3d919", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "4f0e0fe42db8f94dc180509629159d88da54bf3ee65304f8beacb98d52acac27", "generation_seconds_batch": 24.736399626010098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2332563400268555, "target_z": 1.9687502180298475, "ce": 6.316492557525635, "genuineness": 4.381826400756836, "identity": 0.7082082629203796, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/889138f80cb5b5041e0e425162f0c434a68f9fd7ef9574622646b98e93b3d919.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "892cb3d0ab1b152284898b7cb3999ed0ae87bca89ae3df3f3b292b07acbad742", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "7e5c19482995c9d46b871d519d73534016233ca3ce5fbb91f71e4cf830d8d067", "generation_seconds_batch": 18.573795247008093, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2332563400268555, "target_z": 1.9687502180298475, "ce": 6.1894097328186035, "genuineness": 4.118232727050781, "identity": 0.5466936826705933, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/892cb3d0ab1b152284898b7cb3999ed0ae87bca89ae3df3f3b292b07acbad742.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "897265a59b483596df02eb9b40103f294bc725cd0ca516f4b01de4c9b2cb889f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0ddf7bf45d6204753af3ab1cd4a4dd419c588ce2e3b55181b31a8c39865d4bfe", "generation_seconds_batch": 26.496915864991024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.885659694671631, "target_z": 1.7265624974044065, "ce": 5.959072113037109, "genuineness": 3.9173994064331055, "identity": 0.5873591899871826, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/897265a59b483596df02eb9b40103f294bc725cd0ca516f4b01de4c9b2cb889f.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "8975b674cec92a6215bbd11ef340f6a8f20a2474387ea55243fc6e183e0819c9", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6e6b8d61d4b9d1595075307f460ad9cd46264396d12e1f4e2c0c0d2ef5a4ea5d", "generation_seconds_batch": 24.19284507096745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.143553733825684, "target_z": 1.9062499896176264, "ce": 6.38797664642334, "genuineness": 3.8420867919921875, "identity": 0.2184188812971115, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/8975b674cec92a6215bbd11ef340f6a8f20a2474387ea55243fc6e183e0819c9.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "8a75a9f855e7976b442436d91ce0d03a73f491231b09439c5ec882f038fd8aa3", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2cd0fe9d7ec1a2fba7d71ff862414fc50e5d28a90b775619725337b92150ae2d", "generation_seconds_batch": 27.695405201986432, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5380632877349854, "target_z": 1.4843749428969448, "ce": 6.046441555023193, "genuineness": 4.256305694580078, "identity": 0.5352349281311035, "hypothesis": "La porte est encore ouverte ? J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/8a75a9f855e7976b442436d91ce0d03a73f491231b09439c5ec882f038fd8aa3.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "8aeda8e639a41f8d65e70e1009552371322f4c63ff743b1b4498093fc2872cf0", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "6a9efa396a6357b03a5b764c4ff49733e29ad2bd2c4b2c48c4766a725c83776f", "generation_seconds_batch": 24.120895882952027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8071701526641846, "target_z": 1.671874963661692, "ce": 5.792276382446289, "genuineness": 4.011539936065674, "identity": 0.6034234762191772, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt, sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/8aeda8e639a41f8d65e70e1009552371322f4c63ff743b1b4498093fc2872cf0.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "8c6aa42eb1fe1dffc7b165e06c1f026cabccf461d826f815b24d5b792fa8cd95", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "767dd5b04ed888572bc286e8f580a71c056ec333ea5bc2d9721310346b894923", "generation_seconds_batch": 22.754279759014025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.908085346221924, "target_z": 1.7421875545074619, "ce": 6.1894097328186035, "genuineness": 4.243753433227539, "identity": 0.7459266185760498, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/8c6aa42eb1fe1dffc7b165e06c1f026cabccf461d826f815b24d5b792fa8cd95.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "8c7502ec5aeebb11b869d3e92560130936744963669a9761fbd29f3991553a01", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "9035bccefcdd1bf47ed86ddc34cc10a40b0e3ad9cb840f19bb5113c603316f26", "generation_seconds_batch": 23.517710080952384, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.59206485748291, "target_z": 2.2187498027348997, "ce": 6.372091293334961, "genuineness": 3.8671910762786865, "identity": 0.3291042447090149, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/8c7502ec5aeebb11b869d3e92560130936744963669a9761fbd29f3991553a01.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "8c9612f0d535a5b991d7d75396e0f3dd0036ab0a2f5352c5c2036c5b6756eaac", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "62f20bf9224f02080ebce25f3bd1b59923631f28192724c570325f3681e211ee", "generation_seconds_batch": 23.904106890084222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5156378746032715, "target_z": 1.4687500519118684, "ce": 6.213237762451172, "genuineness": 3.8922953605651855, "identity": 0.21579425036907196, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/8c9612f0d535a5b991d7d75396e0f3dd0036ab0a2f5352c5c2036c5b6756eaac.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "8c9d10e884d31ecdc352fcdbbb41a0029ea73d153676afdfd14d06cd8c7721b4", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "e29cba7d0ad77cdb556a4b3962547d9e594c9dad0e75eea99c59d74ba48cd56d", "generation_seconds_batch": 17.47214636905119, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.840808629989624, "target_z": 1.695312549316275, "ce": 6.157639026641846, "genuineness": 4.13078498840332, "identity": 0.555854320526123, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/8c9d10e884d31ecdc352fcdbbb41a0029ea73d153676afdfd14d06cd8c7721b4.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "8ca1c5a3724ad3679c010a19f563e337d4e331bfee3be6445691eb67dbf61036", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "7fb80126bfb735343753b7d46ec237d2a28278db461b504ee6fb0d746977350d", "generation_seconds_batch": 24.055840765009634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5156378746032715, "target_z": 1.4687500519118684, "ce": 6.260893821716309, "genuineness": 4.061748504638672, "identity": 0.7313222289085388, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/8ca1c5a3724ad3679c010a19f563e337d4e331bfee3be6445691eb67dbf61036.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": "8cd6c43c8a03631c5c200214c59e9e1eb418bd35f5f1bdb3ad37818c1ad48605", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0ea303b2862d3c3fa1f2b3ad7fc515ed08933fd718381ea1a4418da3df701dae", "generation_seconds_batch": 16.375146290054545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7398934364318848, "target_z": 1.6249999584705053, "ce": 6.419747352600098, "genuineness": 4.545003414154053, "identity": 0.5539413094520569, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/8cd6c43c8a03631c5c200214c59e9e1eb418bd35f5f1bdb3ad37818c1ad48605.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": "8d02b716dd5eb0244184044c09c4b4bc1cdbf8478725030eae9469cc70abd691", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2b7a6568d2efb006501503666bfac694a5fdce3788c791400ab0b04426326381", "generation_seconds_batch": 21.613740868051536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.268836498260498, "genuineness": 4.174716949462891, "identity": 0.6602257490158081, "hypothesis": "Ik schreef het nummer op de envelop, lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/8d02b716dd5eb0244184044c09c4b4bc1cdbf8478725030eae9469cc70abd691.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "8d54a7a7531a2228e9ad04608063901ae14bdd1728022610bfbe6cf0a5712b5f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "4c3913ab8fee0e966c1df36ab3ac5d895f0db09893e1e52eab5996e6c0199588", "generation_seconds_batch": 25.671361043001525, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.614490509033203, "target_z": 2.2343748598379554, "ce": 6.419747352600098, "genuineness": 4.13078498840332, "identity": 0.6595634818077087, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/8d54a7a7531a2228e9ad04608063901ae14bdd1728022610bfbe6cf0a5712b5f.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "8d97614d1535e784c43297c2fe84087770111c6270f4507412417fc7d6fa9da1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "7e511dad090369c269ef9bab54b216a70d9c9bad851dc5b35c98b4bd8ef7dd27", "generation_seconds_batch": 18.018651907914318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 6.086154937744141, "genuineness": 4.20609712600708, "identity": 0.5750291347503662, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/8d97614d1535e784c43297c2fe84087770111c6270f4507412417fc7d6fa9da1.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "8e6192d36e8b9a064151a1d10ca2123c618949748f6cd17dfbe5d78f2ec91966", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "565ee9efe58397fd1d8f92f957537260831ed3b16eb56a24c5a40dad10e67316", "generation_seconds_batch": 24.907093473011628, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.098702430725098, "target_z": 1.8749998754115158, "ce": 5.919358730316162, "genuineness": 4.20609712600708, "identity": 0.597488284111023, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/8e6192d36e8b9a064151a1d10ca2123c618949748f6cd17dfbe5d78f2ec91966.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "8ea23a1c0fac2ef2022b88e2c381ae56739a608da9db53458d400dbef3e3398c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "62bfad5ad89748dad759e234c79ccc8874ecd4943dfe17cf38c8591c64679697", "generation_seconds_batch": 24.593137993942946, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.053851127624512, "target_z": 1.8437497612054052, "ce": 6.2767791748046875, "genuineness": 3.8922953605651855, "identity": 0.3755536675453186, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/8ea23a1c0fac2ef2022b88e2c381ae56739a608da9db53458d400dbef3e3398c.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "8ea54912c1e05167ff27e71a56b7ebbe140c0bc8e15c4ae531db92e3bb26666b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "a4f7993ab04e9a41c58bcd6a28f675abcf95345a1fc065b169c3971f48961375", "generation_seconds_batch": 27.103716801968403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.229123115539551, "genuineness": 4.18099308013916, "identity": 0.6686347723007202, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/8ea54912c1e05167ff27e71a56b7ebbe140c0bc8e15c4ae531db92e3bb26666b.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": "90028fc9f2d17c221398968aeb08267859e31c59ba8ce1a84a5e53f435127f46", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de3c691132988eec613f09a1ccb4b467f834c6773f8965dc29003e950e26707c", "generation_seconds_batch": 21.532504761009477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.345383644104004, "target_z": 2.0468748390732077, "ce": 6.125868320465088, "genuineness": 4.381826400756836, "identity": 0.7804000973701477, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/90028fc9f2d17c221398968aeb08267859e31c59ba8ce1a84a5e53f435127f46.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "90795ea14daaf0e6e860dcfab6dbe796448833f473890543c1c76d97fdfbf23d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f0ecdb6404d4d6bf5f173bf3ce11dbf1a7cb31a3f1b374a5d149ca3cdba8bd92", "generation_seconds_batch": 24.37193830590695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.009000301361084, "target_z": 1.8124999792352525, "ce": 6.117925643920898, "genuineness": 4.105680465698242, "identity": 0.6556414365768433, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/90795ea14daaf0e6e860dcfab6dbe796448833f473890543c1c76d97fdfbf23d.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": "90e73c263ace81cd5cf5681e94209d3257d5b59da5b576471e2345a03e231dcb", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I wrote the number on the envelope.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9e82fcdcebb55ec1028e404249d7cce00e87934edfd01e76165d0b8b83caa2aa", "generation_seconds_batch": 22.267951305024326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165979385375977, "target_z": 1.9218750467206815, "ce": 6.459460258483887, "genuineness": 3.8797430992126465, "identity": 0.19643840193748474, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/90e73c263ace81cd5cf5681e94209d3257d5b59da5b576471e2345a03e231dcb.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "90ebfd80f68e06363ca66eca80530e3da0b02fa5bcda2bc179b4826c59728114", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "f5036fcfaf6f3f812495eb60bb2257162aeabd5fb9db0bdf5ee1ac6a450a003b", "generation_seconds_batch": 24.38863174605649, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.009000301361084, "target_z": 1.8124999792352525, "ce": 5.982900142669678, "genuineness": 4.293961524963379, "identity": 0.569126307964325, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce calcul.", "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_S_NARR_reg/90ebfd80f68e06363ca66eca80530e3da0b02fa5bcda2bc179b4826c59728114.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": "90ee1f9d5c629db02017661fb60ad5ea8b1812cde78c313452fb8ab6cc67b362", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca402d808a65fd76f51dc5aee74d71f0b152433ca87af57bd8d6b5c6f3cd2085", "generation_seconds_batch": 21.010833706939593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.173524379730225, "genuineness": 4.281409740447998, "identity": 0.6281797885894775, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/90ee1f9d5c629db02017661fb60ad5ea8b1812cde78c313452fb8ab6cc67b362.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "917d812066215126770f1bcc8b2e356cd281b46f7444ac6d33ed84909c6ee2c2", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "0b3095a31d4ea04b3c4860ff08b6a6970d8416beb89e0a33c0a13fb7a35a23b8", "generation_seconds_batch": 23.603207567008212, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.986574649810791, "target_z": 1.7968749221321973, "ce": 6.292664527893066, "genuineness": 4.030367851257324, "identity": 0.388508141040802, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/917d812066215126770f1bcc8b2e356cd281b46f7444ac6d33ed84909c6ee2c2.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "91e3c1200f437dc4cef14f27f5900cd0ca44ff83346896cfeb59c2832c1ad380", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "5b9dcfd3920b1bb66426c51d139ca3be8ed5f2e1d8f3cd3d06c75cfe87e5c933", "generation_seconds_batch": 24.19284507096745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9977874755859375, "target_z": 1.804687450683725, "ce": 6.1894097328186035, "genuineness": 3.760498285293579, "identity": 0.29238736629486084, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/91e3c1200f437dc4cef14f27f5900cd0ca44ff83346896cfeb59c2832c1ad380.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "927d5e1d29edf5d9fa8292040226b59a8aa87ab5031a8caada9af3837612cbf3", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "fc13ee140b5622651cae71156513248225a92505eb1acf906fc3aebc02dd108c", "generation_seconds_batch": 17.786695753922686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7398934364318848, "target_z": 1.6249999584705053, "ce": 6.252951145172119, "genuineness": 4.2249250411987305, "identity": 0.2848757803440094, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/927d5e1d29edf5d9fa8292040226b59a8aa87ab5031a8caada9af3837612cbf3.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "92abea35ef399b73417cdca461fae6321c857d525975ac17d6a3ff914ed4ec4f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "07e9afb96ab104ebbdb676c84ad75a1b74273d6b436d495cc49d36936737d3b9", "generation_seconds_batch": 17.796968202921562, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.885659694671631, "target_z": 1.7265624974044065, "ce": 5.8875885009765625, "genuineness": 4.068024158477783, "identity": 0.3084065020084381, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro de la casa.", "wer": 0.17647058823529413, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/92abea35ef399b73417cdca461fae6321c857d525975ac17d6a3ff914ed4ec4f.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "935d78605010a876e8e52a6e6d1b5f5ba225d077babd8e3397640d17e3414abd", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bb8673171b23c689f9cf2bec5deb6507dd7897352afbc2f6ee74a8114f542c9e", "generation_seconds_batch": 23.995674683945253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.381084442138672, "target_z": 1.3750000415294947, "ce": 6.070269584655762, "genuineness": 3.7542223930358887, "identity": 0.3791871666908264, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/935d78605010a876e8e52a6e6d1b5f5ba225d077babd8e3397640d17e3414abd.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "9393add9d57e637149a13865bb93cfe009606bbf49744f68272838c57934cf22", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "73826e801654f5e081f74301a7f4274935a4f150ba6769d0476a600c3349a0da", "generation_seconds_batch": 18.644986761035398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3025949001312256, "target_z": 1.3203125077867803, "ce": 6.181467056274414, "genuineness": 3.9927117824554443, "identity": 0.5491564273834229, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in Kijk.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9393add9d57e637149a13865bb93cfe009606bbf49744f68272838c57934cf22.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "940a1a666146a8f6600ee0477d5806d8acb3c89bfc1f0907fddc45613530f88d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "35bd7697a1e788a528c2306e4fb814647d1a3736273bfa0194d7b88f1c9413e0", "generation_seconds_batch": 23.191972568980418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1884050369262695, "target_z": 1.937500103823737, "ce": 6.221180438995361, "genuineness": 4.231201171875, "identity": 0.738199770450592, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/940a1a666146a8f6600ee0477d5806d8acb3c89bfc1f0907fddc45613530f88d.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "945d119afbd1c358764c7da44eb187a4f669b29691ca735aa44a87b67218ffab", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "63cb44f0de4a9907163e10c6d98298ccab8b3db79259394b7a8617540b37ebfd", "generation_seconds_batch": 23.94573297793977, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.659341812133789, "target_z": 2.2656249740440657, "ce": 6.09409761428833, "genuineness": 3.8671910762786865, "identity": 0.6497839689254761, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/945d119afbd1c358764c7da44eb187a4f669b29691ca735aa44a87b67218ffab.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": "94b84a138482de7867a7bce81b98fa85c7ae8218f4f208e00363e4f4cbe46986", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ab6c13a55f25808b8eed57c819bd12ae5025e065a26268be0866cfc67139de5", "generation_seconds_batch": 18.02554661198519, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7398934364318848, "target_z": 1.6249999584705053, "ce": 6.340320587158203, "genuineness": 4.055472373962402, "identity": 0.7371128797531128, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/94b84a138482de7867a7bce81b98fa85c7ae8218f4f208e00363e4f4cbe46986.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "952c32155484f97bdeb6545e240faf33a142f299bc9050f1e9a3970cacf070b7", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "8bdf79442493758812e0332c5f6b9a4b4b3e1b09b557ec761bb78875d19bd746", "generation_seconds_batch": 20.127385213971138, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.502363204956055, "target_z": 2.1562502387945948, "ce": 6.165581703186035, "genuineness": 4.344170093536377, "identity": 0.5639950633049011, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/952c32155484f97bdeb6545e240faf33a142f299bc9050f1e9a3970cacf070b7.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": "95aa8ba0ad2bba6b9755964e3fee502d57fa467f93aaae9c0300286cd27ab62a", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Someone moved the little blue chair.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad7a6373704ac35579b89965d7c80a047e9b32c1cd3cc9c811cf6c86759662f8", "generation_seconds_batch": 22.933450629003346, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.908085346221924, "target_z": 1.7421875545074619, "ce": 6.316492557525635, "genuineness": 4.15588903427124, "identity": 0.36781346797943115, "hypothesis": "Someone moved the little blue chair. It used to stand here beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/95aa8ba0ad2bba6b9755964e3fee502d57fa467f93aaae9c0300286cd27ab62a.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": "960b0a713b0be61dc1454592e7ee0171c83f93178ef64db58f027b5e65fca9f7", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d09676ce5b34c0b6f91756be66f168e2aa5585320fcfb734507c5d36cd7451f", "generation_seconds_batch": 26.433920657029375, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.065064430236816, "target_z": 1.851562621992891, "ce": 6.38003396987915, "genuineness": 4.005263805389404, "identity": 0.5660549402236938, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/960b0a713b0be61dc1454592e7ee0171c83f93178ef64db58f027b5e65fca9f7.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "96a38f84907c39a0ab702b87234d9a4a13c6e7c2d2f28771f7920a12ee76a344", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "7aaf9a6f3042b48d1bb92677bbbd70bb32636bd53291220a9412b61b48e9ab43", "generation_seconds_batch": 22.915816162945703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.087490081787109, "target_z": 1.867187679095946, "ce": 5.998785495758057, "genuineness": 4.268857479095459, "identity": 0.6339783072471619, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/96a38f84907c39a0ab702b87234d9a4a13c6e7c2d2f28771f7920a12ee76a344.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "96a7df845aaf2d60b77c3f52241f040ad5b71bea2f94f18835c78d0c1c224cdc", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "38839f5a680c27768e1d593d01977ac0839f3cee9788fe2324080b2ede8c25fe", "generation_seconds_batch": 22.79389266099315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9977874755859375, "target_z": 1.804687450683725, "ce": 6.197352409362793, "genuineness": 4.124508857727051, "identity": 0.22040805220603943, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/96a7df845aaf2d60b77c3f52241f040ad5b71bea2f94f18835c78d0c1c224cdc.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "97c624ee338efdd43d407eb184ff998e1d680085502479ed48f53cad5388eca7", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 129, "seconds_generated": 10.32, "mp3_sha256": "c0c873b3aded07d3b4f48487121cdd5ac03343b6723bd3e336ba121c753008a0", "generation_seconds_batch": 24.43453179509379, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5941271781921387, "target_z": 1.5234374195366038, "ce": 5.77639102935791, "genuineness": 3.9927117824554443, "identity": 0.5854487419128418, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/97c624ee338efdd43d407eb184ff998e1d680085502479ed48f53cad5388eca7.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "97d3f9de630ba7c31775bdf07da0485587eb51898ebb81a69aa5f20fd91dfd53", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "95e9cd4ca1c21936627bdef5d8397b5d979d5b871e79dad921164e68f28bbdc4", "generation_seconds_batch": 27.649475675076246, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.863234043121338, "target_z": 1.7109374403013513, "ce": 5.7406487464904785, "genuineness": 3.5063188076019287, "identity": 0.5352481603622437, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo está.", "wer": 0.13333333333333333, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/97d3f9de630ba7c31775bdf07da0485587eb51898ebb81a69aa5f20fd91dfd53.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "986ed3d2836190b6f743e4c8ca3bb1f3a99061da9bc1fe039b5d3474d63d23f1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "e67bbbf65362359afb8e993d6d1b9d9479680a4f2c78e7b0b3bb0eaa852ce0fc", "generation_seconds_batch": 29.972136833006516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.292664527893066, "genuineness": 4.3504462242126465, "identity": 0.6376692056655884, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/986ed3d2836190b6f743e4c8ca3bb1f3a99061da9bc1fe039b5d3474d63d23f1.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "98a814ac695194a8d9081c76402df39ae70392cdd8fd17a4bbfa9d9b9592ae5f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "1181afef708db4b00dbaf28c7bc9c58cfb9b73d422891b9726d14b6b7dc8a956", "generation_seconds_batch": 24.37181652802974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.098702430725098, "target_z": 1.8749998754115158, "ce": 5.919358730316162, "genuineness": 4.13078498840332, "identity": 0.4985100030899048, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/98a814ac695194a8d9081c76402df39ae70392cdd8fd17a4bbfa9d9b9592ae5f.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "98bc875594d64bbba1f3fd4366a0768cc429076358fb57c1e418a522fe04cbf2", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e93ed5a04a4ee7e53df12a9c29f222a4aecb650d8e0c5da69da34c0873af0282", "generation_seconds_batch": 17.145667833043262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.076276779174805, "target_z": 1.8593748183084604, "ce": 5.95112943649292, "genuineness": 4.011539936065674, "identity": 0.4035899043083191, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/98bc875594d64bbba1f3fd4366a0768cc429076358fb57c1e418a522fe04cbf2.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "98cc970bb7df2789da0c26f8e1190bddc4d6753e16fd290040b281917eace319", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "45dbb2f385109d92de6f4d7f6609226a9600b1fd6a6cbc00672a5a0774c3e6a2", "generation_seconds_batch": 23.605451787938364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.009000301361084, "target_z": 1.8124999792352525, "ce": 6.316492557525635, "genuineness": 4.168440818786621, "identity": 0.7215845584869385, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/98cc970bb7df2789da0c26f8e1190bddc4d6753e16fd290040b281917eace319.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "98e1f218b8f665ce25384801e59be8ff5ff209a2a7a7c149b9c4cca7994cc42b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "0129bf2ae56019239602755264e22e38861e4afd7f012108bc4142ca344473da", "generation_seconds_batch": 24.28637011698447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.885659694671631, "target_z": 1.7265624974044065, "ce": 6.356205940246582, "genuineness": 4.337893962860107, "identity": 0.5254292488098145, "hypothesis": "I found the little box. It was beside the old window. We can look inside together,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/98e1f218b8f665ce25384801e59be8ff5ff209a2a7a7c149b9c4cca7994cc42b.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "992872d9ac25371796b238f49e18563d4c287d7e5d03770ea95fc3a189d67906", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 127, "seconds_generated": 10.16, "mp3_sha256": "6bcb21224c962b7b64a3eb384ced5b24ab9d5d4270cdb94677071b3058350199", "generation_seconds_batch": 17.092709108954296, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.332377910614014, "genuineness": 3.653805732727051, "identity": 0.16349051892757416, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/992872d9ac25371796b238f49e18563d4c287d7e5d03770ea95fc3a189d67906.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": "99b4a04e53b2f5245ebdd42ff4cfb1f0864afb1803e8d5724b8f44b523635b63", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9184b40f41914fb63e4d11bc5a1583ec415fec7cba67fc5b9c9a3d85ae19d95b", "generation_seconds_batch": 23.256196181988344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.672616720199585, "target_z": 1.5781249532793185, "ce": 6.435632705688477, "genuineness": 4.394378662109375, "identity": 0.7133236527442932, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/99b4a04e53b2f5245ebdd42ff4cfb1f0864afb1803e8d5724b8f44b523635b63.mp3" }, { "id": "heldout-de-0", "language": "de", "sentences": [ "Deine Nachricht kam, als ich schon weg war.", "Ich las sie zweimal... dann legte ich das Handy weg.", "Sag mir, wann wir reden können." ], "text": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal... dann legte ich das Handy weg. Sag mir, wann wir reden können.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-0", "key": "99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a05612a1998667ed189b65aa2f2af4d0921dcc7877dc4a0647a7ee9305bb55ae", "generation_seconds_batch": 20.248717847978696, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4147226810455322, "target_z": 1.3984374610660988, "ce": 6.348263263702393, "genuineness": 4.482243061065674, "identity": 0.5643621683120728, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "99dd1dec2dfbafafa41997a1ee964ead852cb96142e687f3f2e6e75754bc3a92", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "bd3a0b4bb4da917daa39673730f84d2a1ef8f2ca38740a1c367afd53223da971", "generation_seconds_batch": 24.512330768979155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.109915733337402, "target_z": 1.8828127361990015, "ce": 6.197352409362793, "genuineness": 4.168440818786621, "identity": 0.6676182150840759, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/99dd1dec2dfbafafa41997a1ee964ead852cb96142e687f3f2e6e75754bc3a92.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": "9b1cdf8d596db4700587d7cc6c66992e68719014cd051db786c7d094ab09add4", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "67409ca308c6e1ffed6b2a92ceb5935f32039c3d153ee630d4464afda18fabfb", "generation_seconds_batch": 21.08724651299417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.39023494720459, "target_z": 2.0781249532793185, "ce": 6.181467056274414, "genuineness": 4.231201171875, "identity": 0.6813615560531616, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9b1cdf8d596db4700587d7cc6c66992e68719014cd051db786c7d094ab09add4.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": "9b26276bf14f0ac3f4efa75c224147ec3ea6531de6f6dd143fb0f2163406f1fa", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "237a485a4426986a64519b8c1dd7cb7c35118e71aa5fb1b469d97926ee91a7ad", "generation_seconds_batch": 22.58912706200499, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.964149236679077, "target_z": 1.781250031147121, "ce": 6.403861999511719, "genuineness": 4.042920112609863, "identity": 0.48343324661254883, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9b26276bf14f0ac3f4efa75c224147ec3ea6531de6f6dd143fb0f2163406f1fa.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "9b7c791df9b1fc758746e8b6941198915d73bce3c256119022be6ef2c2e630d9", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e9afd6852a4e61e0e49f10fb87b1988075c3ef858f6fafd2559cabaff4d86d3f", "generation_seconds_batch": 24.28637011698447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8520214557647705, "target_z": 1.7031250778678027, "ce": 6.24500846862793, "genuineness": 4.005263805389404, "identity": 0.4043477177619934, "hypothesis": "I found 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_S_NARR_reg/9b7c791df9b1fc758746e8b6941198915d73bce3c256119022be6ef2c2e630d9.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "9b87edf87374732c296928da745027a64d6c430337c86c2095b02ccb1c66bc94", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "61d5f53b530afc291103475cd3e1fc2f3d57f074431bbe60041f290a0b8a70e2", "generation_seconds_batch": 24.901143931085244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.222043037414551, "target_z": 1.9609373572423618, "ce": 5.927301406860352, "genuineness": 4.005263805389404, "identity": 0.4768714904785156, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9b87edf87374732c296928da745027a64d6c430337c86c2095b02ccb1c66bc94.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "9bc471a57597e5243e47d4b133932797ded1abc65ca7c4030f4cd03249be9865", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "e8e4693da41c899f066ca52705468ca7b9add69d9fcd175eb14e6eaad1f954d3", "generation_seconds_batch": 21.31751021707896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.222043037414551, "target_z": 1.9609373572423618, "ce": 6.221180438995361, "genuineness": 4.011539936065674, "identity": 0.5281791687011719, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9bc471a57597e5243e47d4b133932797ded1abc65ca7c4030f4cd03249be9865.mp3" }, { "id": "heldout-nl-3", "language": "nl", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-3", "key": "9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bb8b5583c17b6c865d0b0aae11506f7507a872f9c1794c243c1891cebdf3454a", "generation_seconds_batch": 18.55269476689864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5604889392852783, "target_z": 1.5, "ce": 6.1020402908325195, "genuineness": 4.319066047668457, "identity": 0.5916218757629395, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het naar de lunch terug.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "9c2a81020fd794071b9f3394d4543e8e8c40f3f305e5f185886c87893e6d58ae", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "5e44649f5a3cfc7e5be2bc81c94518f4807fd527660d16da1217d10e643aafda", "generation_seconds_batch": 24.078634591889568, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.659341812133789, "target_z": 2.2656249740440657, "ce": 6.157639026641846, "genuineness": 4.268857479095459, "identity": 0.736228346824646, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9c2a81020fd794071b9f3394d4543e8e8c40f3f305e5f185886c87893e6d58ae.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "94829355a093e33fd0f965ec202d9fd859df349afb72f40352e046511c5f0368", "generation_seconds_batch": 17.31603498896584, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0110626220703125, "target_z": 1.1171875960369566, "ce": 5.665193557739258, "genuineness": 4.607763767242432, "identity": 0.8048132658004761, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "9d33fe27baa6e762599da668b9b1b175c94d1fa4792fbdc02a21b1d0f11a1145", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "fb2e6156fbaafc43f3d6b4eaae73bc9925adcd8e634c17b39aa681820ac915c5", "generation_seconds_batch": 24.186890079989098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.457511901855469, "target_z": 2.1250001245884844, "ce": 6.062326908111572, "genuineness": 3.961331605911255, "identity": 0.7173346877098083, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9d33fe27baa6e762599da668b9b1b175c94d1fa4792fbdc02a21b1d0f11a1145.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "9d4707299b54e2227faccf826a8d78616da36f389a702fcbffb51a6c2e7c2c90", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c121381e167e5927f163978460ee9218d9988bbe28b05bf2935b33c2a55376c0", "generation_seconds_batch": 23.904106890084222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8652963638305664, "target_z": 1.0156250571030552, "ce": 6.141753673553467, "genuineness": 3.873466968536377, "identity": 0.3721253573894501, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9d4707299b54e2227faccf826a8d78616da36f389a702fcbffb51a6c2e7c2c90.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "9d504377b27742536cd5814f2e574b743ed49afe407ac9a587e0a353f2b997d6", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "ddfde40b99220c66d5530bbd3a149b9d66b11eeb04a9a3b39063d98717376178", "generation_seconds_batch": 24.260009414050728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8744468688964844, "target_z": 1.718749968852879, "ce": 6.197352409362793, "genuineness": 4.400654315948486, "identity": 0.48167911171913147, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle comptait.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9d504377b27742536cd5814f2e574b743ed49afe407ac9a587e0a353f2b997d6.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "9da22fe3837467327b111e6fc17d8413a478d3fcbf647cf93246ef99517afe40", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "b75c24be628008667c3e9f460c0592f7767dce99f02a20cb68253c02241ea81d", "generation_seconds_batch": 23.39317159296479, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.087490081787109, "target_z": 1.867187679095946, "ce": 6.213237762451172, "genuineness": 4.149612903594971, "identity": 0.7763708829879761, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9da22fe3837467327b111e6fc17d8413a478d3fcbf647cf93246ef99517afe40.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "9ddacdfdce76c151c8d20034f7920023dd825f26b7d4b8a08af870dac074a474", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "eb64f4e10f4925e664deb6552363b3dd3db53ff2fe667a22118db7ef082900c1", "generation_seconds_batch": 23.45346722495742, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.986574649810791, "target_z": 1.7968749221321973, "ce": 6.197352409362793, "genuineness": 4.319066047668457, "identity": 0.7417702674865723, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9ddacdfdce76c151c8d20034f7920023dd825f26b7d4b8a08af870dac074a474.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "9e00601cff9618bb18e44954056065bc1984c288e5638416dc27df232c9d5815", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "ddace2d4c017da0600a97bfd4193e6393e22047170d636c6671699b96dd369d2", "generation_seconds_batch": 21.411808006931096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.730742931365967, "target_z": 0.9218750467206815, "ce": 5.649308204650879, "genuineness": 3.8420867919921875, "identity": 0.19386053085327148, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9e00601cff9618bb18e44954056065bc1984c288e5638416dc27df232c9d5815.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "9e548ba645a0a478caa22ddb2ed875bca641a6cac7484a23e14016b6b9e80ebb", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "ce4c2ff21ce5240a70819f082e4bd85d93e75e783f359211281a21fc55e81e62", "generation_seconds_batch": 18.6391249129083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7286806106567383, "target_z": 1.6171874299189777, "ce": 6.252951145172119, "genuineness": 4.055472373962402, "identity": 0.17359046638011932, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9e548ba645a0a478caa22ddb2ed875bca641a6cac7484a23e14016b6b9e80ebb.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "9e8d462cf262132e2365d1aeed363e6bbd9ee2f2665a13c3d8a8ca6229f41367", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "553cb6002fb62dbf8c310fb0bbf0cd46935adaa5775c648267d818fbab37b3c2", "generation_seconds_batch": 23.998685345984995, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.636916160583496, "target_z": 2.2499999169410105, "ce": 6.0146708488464355, "genuineness": 4.068024158477783, "identity": 0.4992508888244629, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9e8d462cf262132e2365d1aeed363e6bbd9ee2f2665a13c3d8a8ca6229f41367.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "9f74f9803699b297bd4f2abafcd4827134c2bffdc7b8f0c628d6ae5822896499", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "66455748814af57edbb8277998e3e05ac935d36bda446b953cb3642de98434ae", "generation_seconds_batch": 23.928455842891708, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3250203132629395, "target_z": 1.3359373987718566, "ce": 5.919358730316162, "genuineness": 3.8671910762786865, "identity": 0.13575947284698486, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9f74f9803699b297bd4f2abafcd4827134c2bffdc7b8f0c628d6ae5822896499.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "9fc3a45f1bae426bad9e51c137e3863fd0ca1876dd62603e362f9984d67d4bcd", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9042c25640f0bb5fb36a8a648b0109cd6c4277cf2e4d8a3afb29c64f8f4580bb", "generation_seconds_batch": 16.489074416924268, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.481999397277832, "target_z": 1.4453124662572856, "ce": 6.117925643920898, "genuineness": 3.9927117824554443, "identity": 0.11332175135612488, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9fc3a45f1bae426bad9e51c137e3863fd0ca1876dd62603e362f9984d67d4bcd.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "9ff3752fb77ec5ba64baad2d8f4c73226507098e0a5791707aaf994d475b81f3", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "07bf25ceee96f1e53e56695a74b958ca86c41c3da2a21197b2c85a5cb4d6ef21", "generation_seconds_batch": 23.878003800986335, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.062326908111572, "genuineness": 3.9299514293670654, "identity": 0.33604711294174194, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/9ff3752fb77ec5ba64baad2d8f4c73226507098e0a5791707aaf994d475b81f3.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "a0043c0ecf27805a6cf6b26716a9635507f5a15027f0f0aa1209c7477fe25d5f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "31341cca02f8851d0eb42b61ddda245267f8e3fe919e78a126665f87f982a047", "generation_seconds_batch": 24.634088345919736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.076276779174805, "target_z": 1.8593748183084604, "ce": 5.982900142669678, "genuineness": 3.8420867919921875, "identity": 0.3546164631843567, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a0043c0ecf27805a6cf6b26716a9635507f5a15027f0f0aa1209c7477fe25d5f.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "a06e74c791c4930f8404c5fe7c9a5efab0100781bdb852eb6ef029bb21051e61", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "b509ef1f33440c5266b3f81d5d7eb0228a031de531a7129f2272f858c1f8ea5d", "generation_seconds_batch": 23.852780757006258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2108306884765625, "target_z": 1.9531251609267921, "ce": 6.284721851348877, "genuineness": 4.20609712600708, "identity": 0.3687010407447815, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a06e74c791c4930f8404c5fe7c9a5efab0100781bdb852eb6ef029bb21051e61.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "a093c971d0051c90bf8569dde31ec8e09533e356e55ee069d4a480a1c5305660", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "41d190032e3ad57ad1eef346435d45d9e61f0ea7258f0e0a20f449d339bb8cc3", "generation_seconds_batch": 23.647414525970817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.109915733337402, "target_z": 1.8828127361990015, "ce": 6.141753673553467, "genuineness": 3.7040140628814697, "identity": 0.11732757091522217, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a093c971d0051c90bf8569dde31ec8e09533e356e55ee069d4a480a1c5305660.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "a0a45e0ce92c45a99983d288bd47bb8eafbd39680855b125e3f96e83ef1c47e4", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "4a07bc4edac09c9027b8d00b7b2482c3b7aa4fab4319ae5fd18159db0e70449c", "generation_seconds_batch": 24.357175340061076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.627765655517578, "target_z": 1.5468750051911868, "ce": 6.372091293334961, "genuineness": 4.068024158477783, "identity": 0.28569459915161133, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a0a45e0ce92c45a99983d288bd47bb8eafbd39680855b125e3f96e83ef1c47e4.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "a0a7b3e9cb4a690eb9ca48132e9fa7d63e3cd778717dbd96efc141636fdab4be", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "9b3247bf3c5d7c64798d3cca6c46e9dc8a5b8cc55fcaad60ae12729b80d03959", "generation_seconds_batch": 24.738116881926544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1884050369262695, "target_z": 1.937500103823737, "ce": 6.197352409362793, "genuineness": 4.118232727050781, "identity": 0.6898790597915649, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a0a7b3e9cb4a690eb9ca48132e9fa7d63e3cd778717dbd96efc141636fdab4be.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "a0f5618362cff4d489d30b280aa02faad0e46bd46b257fdd5a06e4f190de4cc7", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "ecb77f1e7bb7b9d25b05ef632888f79b202619cafa2531a863c14719900761ac", "generation_seconds_batch": 24.205523965065368, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.704193115234375, "target_z": 2.2968750882501765, "ce": 6.403861999511719, "genuineness": 3.8044304847717285, "identity": 0.2636524438858032, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a0f5618362cff4d489d30b280aa02faad0e46bd46b257fdd5a06e4f190de4cc7.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "a1123ade330e3fd5ee8eac21ca40e2eba30cdbdf704e0134c25f48496d4d6efc", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "610f7264c47ab396a7ae26b17298dc12f8e9290f96f8e72c54cee7ea31ac2669", "generation_seconds_batch": 24.423882469069213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9437856674194336, "target_z": 1.0703124247277909, "ce": 5.756534099578857, "genuineness": 3.7416701316833496, "identity": 0.5447671413421631, "hypothesis": "Zol de deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a1123ade330e3fd5ee8eac21ca40e2eba30cdbdf704e0134c25f48496d4d6efc.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "a11c1bfe632f82d5abd5efcbf6f87c2291805367368f35919cb16cd1cb373218", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "502132a30e6237ddc4edea7b222d4609cc511cd757f134f5550bd3250d8254ea", "generation_seconds_batch": 24.23634832899552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7847447395324707, "target_z": 1.6562500726766158, "ce": 6.38003396987915, "genuineness": 4.432034492492676, "identity": 0.5000552535057068, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a11c1bfe632f82d5abd5efcbf6f87c2291805367368f35919cb16cd1cb373218.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": "a2c274d5cd75c5845934f24f541f693dc65e8f3c0057704c8ef009c6120073a7", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "73aa26765d07c4bcb07a2e3b86865cd8f35408ae6f9403a85132d9fe79e53961", "generation_seconds_batch": 21.613740868051536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.266894340515137, "target_z": 1.9921874714484724, "ce": 6.213237762451172, "genuineness": 4.218649387359619, "identity": 0.7697672843933105, "hypothesis": "Ik schreef het nummer op de enveloop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a2c274d5cd75c5845934f24f541f693dc65e8f3c0057704c8ef009c6120073a7.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": "a34a4c373d4385d937932ad8273150a97a934b7bc531755c2d1a04f371613b0b", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7356b5a174c8a723bf089b7f563b6033e98bb2a91370f6f4ad8d3c3c007cffe7", "generation_seconds_batch": 20.733893332071602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.457511901855469, "target_z": 2.1250001245884844, "ce": 6.205295085906982, "genuineness": 4.168440818786621, "identity": 0.6409056782722473, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a34a4c373d4385d937932ad8273150a97a934b7bc531755c2d1a04f371613b0b.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "a36d820e4ee2ae5ec5642b83538457ffece330ae6a4cc116b2971dbb1e8baeb3", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5a24319842b60a5f77be395d60ab4e4386f3c928ab06258c802f7ff2eee7a7f1", "generation_seconds_batch": 20.851321346941404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5829145908355713, "target_z": 1.5156250571030552, "ce": 5.665193557739258, "genuineness": 3.8044304847717285, "identity": 0.256599485874176, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a36d820e4ee2ae5ec5642b83538457ffece330ae6a4cc116b2971dbb1e8baeb3.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "a3b5b799e0cc08c401608cea9d3bff95e0818ce3fa67b4596a8e1f79712325d5", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "0714836d39cef89add23d19c07bacf0a96c34bedd911d0f126e35c7c57a3c771", "generation_seconds_batch": 26.436338186031207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.076276779174805, "target_z": 1.8593748183084604, "ce": 6.125868320465088, "genuineness": 4.287685871124268, "identity": 0.644061803817749, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a3b5b799e0cc08c401608cea9d3bff95e0818ce3fa67b4596a8e1f79712325d5.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "a43bd70840e507de6ccdb22832e935dee06b0602eb4453083a6fa555e6c0ae63", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "a5a7d3654e0e5d31ffe46b4ed7af040e2541994d51a081d0d9103735436f5a37", "generation_seconds_batch": 18.12809932802338, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2577438354492188, "target_z": 1.2890625596986487, "ce": 6.133810997009277, "genuineness": 3.986435890197754, "identity": 0.36916881799697876, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a43bd70840e507de6ccdb22832e935dee06b0602eb4453083a6fa555e6c0ae63.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "a48b9c176b94d58379dfdfd7d061283a9f5d22f449d269776bc1b1ea8c183ccd", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "782053e803c14de06746d4fc48d009e85183b9dd6f8cfb07ab70e70feccf0473", "generation_seconds_batch": 30.63822686392814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.39023494720459, "target_z": 2.0781249532793185, "ce": 5.863759994506836, "genuineness": 3.9048471450805664, "identity": 0.45994123816490173, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a48b9c176b94d58379dfdfd7d061283a9f5d22f449d269776bc1b1ea8c183ccd.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "a49068d0e03e7305711da3a0a4ef823b5590f47ea140f593679a1e44fc582eb5", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "078a66a05725f7e5553b7e48b316e3929b18f1c6c2cf5305ddf9f6900ddb473a", "generation_seconds_batch": 23.836258416064084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.121128082275391, "target_z": 1.890624932514571, "ce": 6.125868320465088, "genuineness": 4.042920112609863, "identity": 0.6803180575370789, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a49068d0e03e7305711da3a0a4ef823b5590f47ea140f593679a1e44fc582eb5.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "a4bbac0f1f6792c7d965a3af96b5ee114f58f51ea4d7608dccae4c885324b4c6", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "edb23075417912bef4c39fba7ff63ed52380afe4dc9672616e1f0165ac91f969", "generation_seconds_batch": 23.143791874987073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.403510093688965, "target_z": 1.3906250986325501, "ce": 6.054384231567383, "genuineness": 4.293961524963379, "identity": 0.8373347520828247, "hypothesis": "La porte y a y a encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.2222222222222222, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a4bbac0f1f6792c7d965a3af96b5ee114f58f51ea4d7608dccae4c885324b4c6.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "a5b3bdd7599774f53addae48acd466bf52355df8564f5e4bfff1518da12df3db", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c12f4932b9aaedc0efa0f4cd0f3693d66641ddaffcd780cbd4848124960fac1b", "generation_seconds_batch": 18.334531052038074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4707865715026855, "target_z": 1.4374999377057578, "ce": 6.133810997009277, "genuineness": 3.911123275756836, "identity": 0.8438878059387207, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a5b3bdd7599774f53addae48acd466bf52355df8564f5e4bfff1518da12df3db.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "a6130564125196ca748319f67ce0b7bb8cb4fe2b31fc4af2b0ec87dec3197238", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "4f5fb4c17017306beb7214ccbea6d29c1ae1563089964e97c23d1314f898a994", "generation_seconds_batch": 24.515246671973728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.345383644104004, "target_z": 2.0468748390732077, "ce": 6.078212261199951, "genuineness": 4.093128204345703, "identity": 0.4160712957382202, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a6130564125196ca748319f67ce0b7bb8cb4fe2b31fc4af2b0ec87dec3197238.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "a647c76c6ae96a95855a66b67820b87d04598972d6d158cc978f746e53cd3ead", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "922389e3829fc880dda3437365f917513706725235e2cc546d9a09713f30ed41", "generation_seconds_batch": 22.18837694893591, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.313807725906372, "target_z": 1.328125036338308, "ce": 6.133810997009277, "genuineness": 3.7918784618377686, "identity": 0.35568198561668396, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a647c76c6ae96a95855a66b67820b87d04598972d6d158cc978f746e53cd3ead.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": "a659bbcafd5bf4ab90ffc102af584f6b9217da20ab75bb92e6c827742c8483fc", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Your message arrived after I had left.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57f4d778806ef9d747aaa95ee0efe8f081b70de0d9c7ed5cf7530dc62278d0ee", "generation_seconds_batch": 24.946528493077494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2689566612243652, "target_z": 1.2968750882501763, "ce": 6.38797664642334, "genuineness": 3.8420867919921875, "identity": 0.3883245587348938, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a659bbcafd5bf4ab90ffc102af584f6b9217da20ab75bb92e6c827742c8483fc.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "a6986dee480de218fa7f0d30510dc76aa45aaf9c4852e86cf49d93fb51e7b90f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4dc24674b6ad2901d5927d326012c3710bf6f28a9c37acd14c83b6bd5a803db6", "generation_seconds_batch": 23.06795462197624, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.481999397277832, "target_z": 1.4453124662572856, "ce": 6.1894097328186035, "genuineness": 3.898571252822876, "identity": 0.3624257743358612, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a6986dee480de218fa7f0d30510dc76aa45aaf9c4852e86cf49d93fb51e7b90f.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "a6a3f68fc421c28e7de4cd4e7178e984a7d8fa58c971dcef7ad36e153992fd0a", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ded27edf343fefacee5eea810c7fc264ce2fb5923e0782ffeb437c9f2137b966", "generation_seconds_batch": 28.71340654406231, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.717468023300171, "target_z": 1.609375067485429, "ce": 6.125868320465088, "genuineness": 3.8546390533447266, "identity": 0.09601783752441406, "hypothesis": "The door is till open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a6a3f68fc421c28e7de4cd4e7178e984a7d8fa58c971dcef7ad36e153992fd0a.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "a6a7820e09877436202e63eecf492c13bb6a03a30605ffc5a2eef19e084a93a5", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "8ae0e12eada4a62a693a50327ac56b39c0261fb896fe076be22d5f8b736f6f8a", "generation_seconds_batch": 21.31751021707896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2332563400268555, "target_z": 1.9687502180298475, "ce": 5.967014789581299, "genuineness": 4.099404335021973, "identity": 0.7357989549636841, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a6a7820e09877436202e63eecf492c13bb6a03a30605ffc5a2eef19e084a93a5.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": "a6b278d642f647b7b98dd33290eac00b6da65bd42d54ca16980f1c5c09022d5d", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "56b6319868472be410dea3a1ff4679f986ea682da818648e6177deecb5692692", "generation_seconds_batch": 18.12534519098699, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.345383644104004, "target_z": 2.0468748390732077, "ce": 6.300607204437256, "genuineness": 4.011539936065674, "identity": 0.5944563150405884, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter, bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a6b278d642f647b7b98dd33290eac00b6da65bd42d54ca16980f1c5c09022d5d.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "a6c060d34102fe2d9449a282dd764eb2eee0d6670634867f0d7ae871dda0864b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7fc8aaa4f3bd049a9908d77dac36d862a8e3169adf3a3351bb01ccce79498fae", "generation_seconds_batch": 17.092709108954296, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.260893821716309, "genuineness": 3.9173994064331055, "identity": 0.2855502665042877, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a6c060d34102fe2d9449a282dd764eb2eee0d6670634867f0d7ae871dda0864b.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "a6e9d8986972f7605e613edb856d09b69e16ec31718d31f9257f4987c1c7a156", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "f48593ab751d6996d861ec0787287d89bcb049138c915054cd29ffce5dd2933f", "generation_seconds_batch": 16.915298811043613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.177192211151123, "target_z": 1.9296875752722091, "ce": 6.125868320465088, "genuineness": 3.7165660858154297, "identity": 0.26087233424186707, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a6e9d8986972f7605e613edb856d09b69e16ec31718d31f9257f4987c1c7a156.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": "a6eac83ffdea3bc4c79f2d1c00972a982e584c00572014fb3a0d345d4e576a5a", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The keys are on the kitchen table.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4246fb1a2c634a2fa93be41f2226f479f661e03bdf68256819f436b5dbd3f915", "generation_seconds_batch": 24.034143017954193, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.009000301361084, "target_z": 1.8124999792352525, "ce": 6.3641486167907715, "genuineness": 4.256305694580078, "identity": 0.3502920866012573, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a6eac83ffdea3bc4c79f2d1c00972a982e584c00572014fb3a0d345d4e576a5a.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "a71e64b12c8dea8962f5d78e0f61177621e816a68bf04c38c26225ea0694f7f4", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9e2c31b14c826321cdf5759cf2d9957f1371f57aeb2fc72b8e274c965ac58287", "generation_seconds_batch": 23.091814556973986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8968725204467773, "target_z": 1.7343750259559343, "ce": 6.181467056274414, "genuineness": 4.093128204345703, "identity": 0.48520880937576294, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a71e64b12c8dea8962f5d78e0f61177621e816a68bf04c38c26225ea0694f7f4.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "a785520505d1b3fe68ad8eb1aeac31a4b901994d655d118ff2ee3498a9e721ff", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "d64ab1caa062831c2a6419c3eddfac9c8367bf5a6ab858f7d6e16816b93ef8b1", "generation_seconds_batch": 23.81718822696712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.999849796295166, "target_z": 1.109375067485429, "ce": 6.252951145172119, "genuineness": 4.049196243286133, "identity": 0.5772128105163574, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a785520505d1b3fe68ad8eb1aeac31a4b901994d655d118ff2ee3498a9e721ff.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "a7ddc40695e4fdc420edc0899af95ede69ae83cfba05e38731aad8cc8a11628c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "898d3e816b42021b0782bd2ccc29d7ba19ff7553f9066dcd35afdfbc5ab3aedc", "generation_seconds_batch": 17.53504059801344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.031425952911377, "target_z": 1.828125036338308, "ce": 6.165581703186035, "genuineness": 3.873466968536377, "identity": 0.3625842034816742, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a7ddc40695e4fdc420edc0899af95ede69ae83cfba05e38731aad8cc8a11628c.mp3" }, { "id": "heldout-nl-4", "language": "nl", "sentences": [ "Ik schreef het nummer op de envelop.", "Lees het goed... en controleer de laatste twee cijfers.", "Bel me wanneer je even tijd hebt." ], "text": "Ik schreef het nummer op de envelop. Lees het goed... en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-4", "key": "a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fbed86815969cd1b3784b00ca93f8c2d7d7a4af6f67797b3f62ab7e2b490f08a", "generation_seconds_batch": 18.40612195606809, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.672616720199585, "target_z": 1.5781249532793185, "ce": 6.062326908111572, "genuineness": 4.15588903427124, "identity": 0.48193082213401794, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "a8e6b782f672f9cc6a0963a5c95b0ac5cb9cae441286a75a5afbee98b3f0407f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "982c407c7080d342da871a5fd9089a3ae896fd189bf8bc0ec48877809c7fab85", "generation_seconds_batch": 24.615514816949144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 6.133810997009277, "genuineness": 4.256305694580078, "identity": 0.6515229940414429, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a8e6b782f672f9cc6a0963a5c95b0ac5cb9cae441286a75a5afbee98b3f0407f.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "a8f90638c9e5202fc8d4b3c1dfb1eb197a154f6e2dda8ccc9b8f38b07ceff668", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "d65393e67593be8ef6ec777c5c2ed729cf44e0dff46359a1964c42685daf2c1f", "generation_seconds_batch": 24.172320017940365, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.278107166290283, "target_z": 2.0, "ce": 6.109982967376709, "genuineness": 3.5816311836242676, "identity": 0.05300294607877731, "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_S_NARR_reg/a8f90638c9e5202fc8d4b3c1dfb1eb197a154f6e2dda8ccc9b8f38b07ceff668.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "a8fd6e1ba7dccba7d161883466100de37e7fbb4f93ee5b977cd429e935570966", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d421b176d1f68c526c262be3209f9885056e39f914470e2b17838d91347ea43c", "generation_seconds_batch": 23.51513467996847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1344032287597656, "target_z": 1.2031250778678027, "ce": 6.125868320465088, "genuineness": 3.6914620399475098, "identity": -0.057030316442251205, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a8fd6e1ba7dccba7d161883466100de37e7fbb4f93ee5b977cd429e935570966.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "a921c583054cb2604f0db5bcdc484b261e2acdc4a12f1caba533c687a7f2fc50", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8bac09df75e5019d5dbd1942851ccc518fd3ee2f2dba7050ae57905665960328", "generation_seconds_batch": 19.41214348305948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5247883796691895, "target_z": 2.171874963661692, "ce": 6.0305562019348145, "genuineness": 4.093128204345703, "identity": 0.7085933089256287, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a921c583054cb2604f0db5bcdc484b261e2acdc4a12f1caba533c687a7f2fc50.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f167059283e463fd40e60f75361476897457bd3c5c04dc24b2af8516ed049049", "generation_seconds_batch": 16.61897993204184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7398934364318848, "target_z": 1.6249999584705053, "ce": 6.133810997009277, "genuineness": 4.532451152801514, "identity": 0.795751690864563, "hypothesis": "Ton message est arrivé après mon départ, je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "a984f19b8554e4c31d4494ed43301b5a7c285ee369943ea104eb32a60f04c8aa", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "81071364bc3c39a2a1d06c695d7da1137c29c2dacd734cddb3cc751fd3d3ddfc", "generation_seconds_batch": 24.7911069910042, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165979385375977, "target_z": 1.9218750467206815, "ce": 6.022613525390625, "genuineness": 4.174716949462891, "identity": 0.43073421716690063, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/a984f19b8554e4c31d4494ed43301b5a7c285ee369943ea104eb32a60f04c8aa.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "aa1bcc55830967d9a693fd32cacdbff3ffa032e40b1a4ba554e613e18bd657a2", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "6a3c85fe21130f0262fcba50551b5a3e228f698437d1220e64617465a545e27b", "generation_seconds_batch": 24.461215593968518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9213602542877197, "target_z": 1.0546875337427144, "ce": 5.673136234283447, "genuineness": 3.8922953605651855, "identity": 0.23228183388710022, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/aa1bcc55830967d9a693fd32cacdbff3ffa032e40b1a4ba554e613e18bd657a2.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": "aaf44ab27a9b7db62fe84d73bcc8bbee0b0d214989db0977cec5811919b6b05b", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ab50095d56cc06d1f85d706437acf02b3228d68580ba6008baaf65eaa0f50fa", "generation_seconds_batch": 18.25528566900175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.39023494720459, "target_z": 2.0781249532793185, "ce": 6.173524379730225, "genuineness": 4.218649387359619, "identity": 0.6818805932998657, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/aaf44ab27a9b7db62fe84d73bcc8bbee0b0d214989db0977cec5811919b6b05b.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "aafc0655d4559e62c5f26ccfc3f6e42148e395b477160cb544274bdb9c5242be", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f3d2cc19ccc7e32b5c8005383f37e28939233666a92535cb95c1a17ef61e97a9", "generation_seconds_batch": 24.260009414050728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25568151473999, "target_z": 1.9843749428969448, "ce": 5.748591423034668, "genuineness": 3.911123275756836, "identity": 0.5088656544685364, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/aafc0655d4559e62c5f26ccfc3f6e42148e395b477160cb544274bdb9c5242be.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "ab3346f20525fd61e4e0058d481b4497b16543d1884e265781631e79a0166ff4", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "b40e3e000635afe7811ee8165bed3abe2973d7156402be8c11216883fddc3d1d", "generation_seconds_batch": 24.023838283959776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.749044418334961, "target_z": 2.328125202456287, "ce": 6.300607204437256, "genuineness": 4.18099308013916, "identity": 0.6613507270812988, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ab3346f20525fd61e4e0058d481b4497b16543d1884e265781631e79a0166ff4.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "ab565e0c95bed8edb6e7346b5ddd393fcc0ec35e97e4e03666472d8372241fcc", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "8d797b724504d603018bc592d978eff1dbd514ad5efc4debd4b7ea568d38fdd9", "generation_seconds_batch": 25.715329775004648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8071701526641846, "target_z": 1.671874963661692, "ce": 6.022613525390625, "genuineness": 4.231201171875, "identity": 0.4851212501525879, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ab565e0c95bed8edb6e7346b5ddd393fcc0ec35e97e4e03666472d8372241fcc.mp3" }, { "id": "heldout-en-0", "language": "en", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-0", "key": "ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "36ee0af955e9ccb1b32f9598d88fbc102325987c55c840fb8a4cf3e9be30caa9", "generation_seconds_batch": 17.811453575035557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.85408353805542, "target_z": 1.0078125285515276, "ce": 6.523001670837402, "genuineness": 4.256305694580078, "identity": 0.37033811211586, "hypothesis": "Your message arrived after I had left. I read it twice before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "ab8af5130a13b259d19d82b9656c0685c1d046811a3d42366fcf61368eb22ff9", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "cc15da6a05db92ea85060bebc796fda963b2118644004527bc765397bb8991f2", "generation_seconds_batch": 24.346117430948652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.818382978439331, "target_z": 1.6796874922132197, "ce": 6.173524379730225, "genuineness": 4.20609712600708, "identity": 0.7503951787948608, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ab8af5130a13b259d19d82b9656c0685c1d046811a3d42366fcf61368eb22ff9.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "ac1177869c5312b8dbb3e04bc7fb92993b3d88f7f2cf039bed6a779af41708cf", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "c3c6651ac5c50f3c8e2135f3f431141474a9708d45afc4f414eeacf0f916eeee", "generation_seconds_batch": 23.735356541001238, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5247883796691895, "target_z": 2.171874963661692, "ce": 6.022613525390625, "genuineness": 4.293961524963379, "identity": 0.5476059913635254, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ac1177869c5312b8dbb3e04bc7fb92993b3d88f7f2cf039bed6a779af41708cf.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "aca3dc8bd3c41b1c2f84352079564eb8412b222df45f2c9b3c33b2c7ee4369fb", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "343665e0b59b221bda76fe92b118b4e4932759f69cdeafa797378aae2ec8b0b8", "generation_seconds_batch": 24.950757495942526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.278107166290283, "target_z": 2.0, "ce": 6.1020402908325195, "genuineness": 4.243753433227539, "identity": 0.6651967763900757, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/aca3dc8bd3c41b1c2f84352079564eb8412b222df45f2c9b3c33b2c7ee4369fb.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "acde2324eb4f3e4120395e29aa192f568165b0b66154a72a7d8d54ebbd953b7a", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "fe511840f657fe23c7e906bcc752db3ac25522c669b14d15e15aa7dba4f5e91c", "generation_seconds_batch": 18.434154095011763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4932122230529785, "target_z": 1.4531249948088132, "ce": 6.435632705688477, "genuineness": 4.545003414154053, "identity": 0.2821769118309021, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/acde2324eb4f3e4120395e29aa192f568165b0b66154a72a7d8d54ebbd953b7a.mp3" }, { "id": "heldout-nl-0", "language": "nl", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-0", "key": "adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d5c2344d2604499a9c614298decd7913fef21e723bfb48f118d47e0e590cd1ac", "generation_seconds_batch": 17.073086285032332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.212892532348633, "target_z": 1.2578124454925381, "ce": 6.173524379730225, "genuineness": 4.337893962860107, "identity": 0.4395303726196289, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "add3ce3c50aad38429bb6e4b73e27693be62d9e03bea4c7b679e414a099dae4f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4815525ddabff0fac79d05d23c099ec8f75fe8dfe68171cf1f6fdb7e13dcbf01", "generation_seconds_batch": 23.02787744498346, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.358658790588379, "target_z": 1.3593749844264396, "ce": 6.181467056274414, "genuineness": 4.218649387359619, "identity": 0.5806801915168762, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/add3ce3c50aad38429bb6e4b73e27693be62d9e03bea4c7b679e414a099dae4f.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "ae26c163fccb872a4ce0479e9dcbce91db572c38d07904e710871b07fffc2ff6", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "b82e2a0b72372a62f4c00c1b7a95f1ebc1fc6dbe0a368e58d24a2f507b83171c", "generation_seconds_batch": 30.844776742975228, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.042638778686523, "target_z": 1.8359375648898355, "ce": 6.348263263702393, "genuineness": 4.325342178344727, "identity": 0.3295748233795166, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ae26c163fccb872a4ce0479e9dcbce91db572c38d07904e710871b07fffc2ff6.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "ae2f9632e4edacdd33840e8697b254868c3c2f8b3523eb41ea7ddc2b76bdb8a6", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "7ff5d01f90e32482a9ece7c9ac311652d46e95cc00bec52bc44f2468408b9a86", "generation_seconds_batch": 24.109801337006502, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25568151473999, "target_z": 1.9843749428969448, "ce": 5.689021587371826, "genuineness": 3.873466968536377, "identity": 0.5473088026046753, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ae2f9632e4edacdd33840e8697b254868c3c2f8b3523eb41ea7ddc2b76bdb8a6.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "426dc6ffe59decfa9316bd0347962fb0cd7f1b5974bbac66d131070c4e88dde0", "generation_seconds_batch": 14.94840498897247, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.504425048828125, "target_z": 1.4609375233603408, "ce": 6.475345611572266, "genuineness": 4.319066047668457, "identity": 0.4602620601654053, "hypothesis": "The road turns left, behind the bridge. Keep going until you reach this small wooden gate. I will meet you on the other side.", "wer": 0.041666666666666664, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "ae9b1f17a93f42901fc250bd71bcf8118fbc75bd5f0a25f03732e6f612d697ee", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "8d59f805e7744aff4467b1bcc706c7efec315b072bc2385f4e7fb71444573f4a", "generation_seconds_batch": 18.05116326804273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.009000301361084, "target_z": 1.8124999792352525, "ce": 6.229123115539551, "genuineness": 4.344170093536377, "identity": 0.6050116419792175, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ae9b1f17a93f42901fc250bd71bcf8118fbc75bd5f0a25f03732e6f612d697ee.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": "aefd1134658447acc6281ef899e1f99534298201c1fd0d8089219818d13f8be5", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Someone moved the little blue chair.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f340407af4ac4bf0dca56fb58b85f6ba80106294844a24dc82c61b762d7759b6", "generation_seconds_batch": 17.902906979899853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.38003396987915, "genuineness": 3.9550557136535645, "identity": 0.5370707511901855, "hypothesis": "Someone moved the little blue chair. It used to stand here, beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/aefd1134658447acc6281ef899e1f99534298201c1fd0d8089219818d13f8be5.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "af3f65bde750b62369df7e053688775fd36f4a76d56a9c7403e1502c428a6a15", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "df9c6eda22165908128e223a19a87191860228fa5310326fd952087bee03e461", "generation_seconds_batch": 18.49656447500456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.177192211151123, "target_z": 1.9296875752722091, "ce": 5.982900142669678, "genuineness": 4.419482707977295, "identity": 0.5427901744842529, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/af3f65bde750b62369df7e053688775fd36f4a76d56a9c7403e1502c428a6a15.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "afb27a7efadb56fd0a2c135960bd2a73fdc23788ed4b34ff24aa4ff348d0c20b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "4633e35f1e6e10bd68c060787ac005da5960f38f3fb90aacdafb3682773fe043", "generation_seconds_batch": 28.242928871070035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.031425952911377, "target_z": 1.828125036338308, "ce": 6.300607204437256, "genuineness": 4.017816066741943, "identity": 0.5406839847564697, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/afb27a7efadb56fd0a2c135960bd2a73fdc23788ed4b34ff24aa4ff348d0c20b.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": "b0123d4ea0187f41b0b526b59daea6171b7ae68b6a89526d84b245044c88e638", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "637f25cdeb6d1c414df2887a42d2b3aaf65b79b681a2490e05d99c6c3be74559", "generation_seconds_batch": 18.065647045965306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.726618766784668, "target_z": 2.3125001453532317, "ce": 5.990842819213867, "genuineness": 4.030367851257324, "identity": 0.4167833924293518, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b0123d4ea0187f41b0b526b59daea6171b7ae68b6a89526d84b245044c88e638.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "b01b6373bd2b978b1feb188ae9bf0ac63695c76b8f97903936e0011c95422ab4", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "453fffa1d9d7cbeb005e71d992edee944642c638292ff9e1cf4125c3984fc7a0", "generation_seconds_batch": 24.67765392397996, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8295958042144775, "target_z": 1.6875000207647475, "ce": 6.062326908111572, "genuineness": 4.15588903427124, "identity": 0.7036759853363037, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b01b6373bd2b978b1feb188ae9bf0ac63695c76b8f97903936e0011c95422ab4.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "b0e326da43127a5bf72b45833de1870502a5d261188218fd179bc3008b356f68", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d6ed9358fcc9bdcc7e2368f10d547f3d34c8a2bbe806523369906669448b59f", "generation_seconds_batch": 23.160408871015534, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5156378746032715, "target_z": 1.4687500519118684, "ce": 6.0305562019348145, "genuineness": 3.7291181087493896, "identity": 0.282493531703949, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b0e326da43127a5bf72b45833de1870502a5d261188218fd179bc3008b356f68.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "b1252891e71cb045f4124d2ce2687dc164408439935a9e2e9a477cde47cd6572", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f4b77811decff73dcc1e64ace556fb936116a6ea6091a2898ae1081351376a72", "generation_seconds_batch": 24.192906149080954, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5829145908355713, "target_z": 1.5156250571030552, "ce": 6.054384231567383, "genuineness": 3.9801597595214844, "identity": 0.6343321800231934, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b1252891e71cb045f4124d2ce2687dc164408439935a9e2e9a477cde47cd6572.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": "b1a6568d2842b0fd7507bba421aab8006d384e2d4c20ca63566c7b7b5530979e", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4babe6fafcd4e90ab7915894323b1f84a0ed3155d468821bcba11d0a25c261c5", "generation_seconds_batch": 24.649570515961386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.547214031219482, "target_z": 2.1875000207647473, "ce": 6.054384231567383, "genuineness": 3.6789097785949707, "identity": 0.47985273599624634, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b1a6568d2842b0fd7507bba421aab8006d384e2d4c20ca63566c7b7b5530979e.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": "b281824148cba8590f6bd32fcabf3b0dbc2b2db63dea293f96dc93f4d1e8970d", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3431da940e1fa905bde4cc8c32a45c7e0f7c44c188ec964dbd98f9f5a6d1013c", "generation_seconds_batch": 23.57263160601724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.726618766784668, "target_z": 2.3125001453532317, "ce": 6.2767791748046875, "genuineness": 3.8546390533447266, "identity": 0.3747262954711914, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b281824148cba8590f6bd32fcabf3b0dbc2b2db63dea293f96dc93f4d1e8970d.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "126d3b56fcce559b6b34e7b619922cb78d684dfb3a4b4c6f87dffb7b5d9b6558", "generation_seconds_batch": 18.100681693991646, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1119775772094727, "target_z": 1.1875000207647475, "ce": 6.2767791748046875, "genuineness": 4.319066047668457, "identity": 0.675682544708252, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "b49ac523525dc1fb32e05b77cd2409c3e00076a90cfff5d59a202c2a9633c924", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5be3640b101517b13e3ddf45802eaa87fbb6e20599b627010c089697ae803b6d", "generation_seconds_batch": 20.754844077979214, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2016797065734863, "target_z": 1.2499999169410105, "ce": 5.855817794799805, "genuineness": 4.030367851257324, "identity": 0.6650608777999878, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b49ac523525dc1fb32e05b77cd2409c3e00076a90cfff5d59a202c2a9633c924.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "b49e8ca40e1da6767d6e370ad0a2e0e0b288e9d95c43a7760cfe9e8aa2e4a346", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "796974541f62911966c0be3d10239c9ee6756b7ebc785e7bc755fd0939296e06", "generation_seconds_batch": 23.494674212066457, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.547214031219482, "target_z": 2.1875000207647473, "ce": 5.831989288330078, "genuineness": 3.9048471450805664, "identity": 0.6456142663955688, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b49e8ca40e1da6767d6e370ad0a2e0e0b288e9d95c43a7760cfe9e8aa2e4a346.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "b50fb520a2533b98ac3108c33b0c7e64367046079c269b6b0e503489fe9be53f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "678229ba1689d08185fd1dd43844bbded67e242915e3961b7b5c6c2abb242ead", "generation_seconds_batch": 25.53700414195191, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5380632877349854, "target_z": 1.4843749428969448, "ce": 6.149696350097656, "genuineness": 4.243753433227539, "identity": 0.6989952325820923, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b50fb520a2533b98ac3108c33b0c7e64367046079c269b6b0e503489fe9be53f.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "b55fb5430fe86a07f8d25424a25670ad48d5292c7b9d931e17ad0f4d5d159a81", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "30b1c64d9b9cd7a4449d33353f942ffdef04eed81886d041b7bba8b359adb739", "generation_seconds_batch": 23.25226094899699, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.076276779174805, "target_z": 1.8593748183084604, "ce": 6.324435234069824, "genuineness": 4.1998209953308105, "identity": 0.503409743309021, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b55fb5430fe86a07f8d25424a25670ad48d5292c7b9d931e17ad0f4d5d159a81.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "b57ee9768869c687baaaebd4f3b5d65008dd7c717fd2081fe6ebb890ee70defc", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "69e5e6e6114c9614acb79bab2c7796f0731b00fe04fab87e7162e873881c7320", "generation_seconds_batch": 23.81718822696712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4147226810455322, "target_z": 1.3984374610660988, "ce": 5.990842819213867, "genuineness": 3.886019229888916, "identity": 0.5749547481536865, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b57ee9768869c687baaaebd4f3b5d65008dd7c717fd2081fe6ebb890ee70defc.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "b676b4ddfd285a9574df4988033b6b930f5fb90a290e1f2d6f976ee8f81ba0c3", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "bc65683da586ae07f5d87f441d09cde0c9487ba1d0ed28f6d2696dea010558d7", "generation_seconds_batch": 17.200668083969504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7623190879821777, "target_z": 1.6406250155735604, "ce": 6.24500846862793, "genuineness": 4.111956596374512, "identity": 0.6505104899406433, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b676b4ddfd285a9574df4988033b6b930f5fb90a290e1f2d6f976ee8f81ba0c3.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "b7049a69eae7c0f7b821da995025063d26936188e458410411569909628ab9df", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "70c6a5191f20cbebb00f0af4135f1ea56a9f0519a4acec850dfbf7f77a5588db", "generation_seconds_batch": 23.020264480030164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3474459648132324, "target_z": 1.3515624558749118, "ce": 6.022613525390625, "genuineness": 4.093128204345703, "identity": 0.3461650311946869, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b7049a69eae7c0f7b821da995025063d26936188e458410411569909628ab9df.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "b715d99d10a77b9bac6363aed7eb3279caccf9f839c0b2759d399a9ed083e8b4", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "22bf34bff643644d098fcea70bb29f64cc3376cb6bbbf65ea3c8a23aceff59c3", "generation_seconds_batch": 24.56844624702353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7286806106567383, "target_z": 1.6171874299189777, "ce": 6.09409761428833, "genuineness": 3.7416701316833496, "identity": 0.1783997118473053, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b715d99d10a77b9bac6363aed7eb3279caccf9f839c0b2759d399a9ed083e8b4.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "b7cb35c4995a59f4de8fb9322a01e981cc30f9412427a1c1241e6c56df53c8bc", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "e62d17ae8c8a3848fb0411f39ad96ee423e2b3a595307146ed8b308462a8013b", "generation_seconds_batch": 20.851321346941404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.143553733825684, "target_z": 1.9062499896176264, "ce": 5.704906940460205, "genuineness": 3.6192874908447266, "identity": 0.4056825041770935, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b7cb35c4995a59f4de8fb9322a01e981cc30f9412427a1c1241e6c56df53c8bc.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "b7f77e96861862b5dd441caeda6fe1d3fe6efc9daf89ce9de6df05d6bbd83d7c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "804ecfc511681a2a983945f2644ba27c5430d9b0eb46b06a3ef9f561a9b71c1d", "generation_seconds_batch": 20.876710272976197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7623190879821777, "target_z": 1.6406250155735604, "ce": 5.919358730316162, "genuineness": 4.231201171875, "identity": 0.5116350650787354, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b7f77e96861862b5dd441caeda6fe1d3fe6efc9daf89ce9de6df05d6bbd83d7c.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "b85543fafd1449c2beed7329d96b63aa3bffa649021ecac693fc75b433f0d1da", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "6f85e08bd320047ee5132d48352a98a97a93ba2761f396cb4d9c7c4d5af41abb", "generation_seconds_batch": 22.370293823070824, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.908085346221924, "target_z": 1.7421875545074619, "ce": 6.09409761428833, "genuineness": 3.835810661315918, "identity": 0.334463894367218, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b85543fafd1449c2beed7329d96b63aa3bffa649021ecac693fc75b433f0d1da.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "b866c628e048ab5b33dc699247d11f95f6ced25814543506f837132b3e2987bf", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "a4b5b73cdc7db9eb46dcbc32aecf8e38b5a2cf4d7248714d7e8ac2f023a5f383", "generation_seconds_batch": 23.19453237298876, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4483611583709717, "target_z": 1.4218750467206815, "ce": 6.292664527893066, "genuineness": 4.356721878051758, "identity": 0.6886317133903503, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b866c628e048ab5b33dc699247d11f95f6ced25814543506f837132b3e2987bf.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "b8be1c0ff60859b31656f00f51ca8266121077da1c562a1d3697bac03b0190ad", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "7a824e726f6fc9ad5a9beb8c2e77cd5cf12b61c1de6e795d1a42833453cf3d3c", "generation_seconds_batch": 16.864902639994398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.053851127624512, "target_z": 1.8437497612054052, "ce": 6.260893821716309, "genuineness": 4.570107460021973, "identity": 0.7307807207107544, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b8be1c0ff60859b31656f00f51ca8266121077da1c562a1d3697bac03b0190ad.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "b8f2619ab961dc1586a96d318459fc761bd479b0642f528b408c11fe15a34380", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "af16474b6df8e9f14a723d8e7d5681c2d499155e4e74ad20d4540178afc9c29f", "generation_seconds_batch": 24.186890079989098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.031425952911377, "target_z": 1.828125036338308, "ce": 6.205295085906982, "genuineness": 4.143336772918701, "identity": 0.6231793761253357, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/b8f2619ab961dc1586a96d318459fc761bd479b0642f528b408c11fe15a34380.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "07c6997378a7a36dd6b40305086410b37dfd8287a06c0a3465ff3f5fbeac6667", "generation_seconds_batch": 18.033019082038663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0202131271362305, "target_z": 1.8203125077867803, "ce": 6.2767791748046875, "genuineness": 4.287685871124268, "identity": 0.6666876077651978, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends les, quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "bb240ee8be29c49b66be474da8e11fd9b00bbe1855eb1364384eab64455d2b19", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "46bd83c8c0c72b38cdcd7375336ffe641cbf9c14a071c486508177b1b1b89e90", "generation_seconds_batch": 24.221068004961126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.278107166290283, "target_z": 2.0, "ce": 6.356205940246582, "genuineness": 4.086852550506592, "identity": 0.49349507689476013, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/bb240ee8be29c49b66be474da8e11fd9b00bbe1855eb1364384eab64455d2b19.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "bb821b548fe50ecde9d366960642e5d047a723bd7e279c2ed3376a5eb3c8561c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "8d9d0b4d531c86ea16ab7512944ce95bf2e31e5c1496393b149512a596999942", "generation_seconds_batch": 18.006340182968415, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.132340908050537, "target_z": 1.8984374610660988, "ce": 6.09409761428833, "genuineness": 3.9927117824554443, "identity": 0.4552633762359619, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/bb821b548fe50ecde9d366960642e5d047a723bd7e279c2ed3376a5eb3c8561c.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "bbab7ddc59c7b1ae2bc91d0c2c58b8ad920a052c85a0bc6864fd18cce625de0a", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a126aa0b103ce55e9cd68d61ead5d27528a371ba8ecb3d229dd603cf8b63d097", "generation_seconds_batch": 23.647414525970817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.435086250305176, "target_z": 2.109375067485429, "ce": 6.284721851348877, "genuineness": 3.9927117824554443, "identity": 0.42521560192108154, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/bbab7ddc59c7b1ae2bc91d0c2c58b8ad920a052c85a0bc6864fd18cce625de0a.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "bbdc30aff9dbd7f3b910181a000dd629acc07bc2d6cc57c22b348a1801a3c19d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "bf6ea48d57c620de5b4318f0d37e4d3307e18c0d5edf60b1c96590f1cdd9bde6", "generation_seconds_batch": 27.675130528979935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6165528297424316, "target_z": 1.5390624766396592, "ce": 6.221180438995361, "genuineness": 3.8169827461242676, "identity": 0.27466797828674316, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/bbdc30aff9dbd7f3b910181a000dd629acc07bc2d6cc57c22b348a1801a3c19d.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "bbf0a8163773f5ebd39415eec912fd7cfa313c19fec044407cbdd5a30b30f801", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "ad97f73ec95e6199dd44f25b78477aa60eb0725aad85323dee809a25624ee301", "generation_seconds_batch": 24.38150791800581, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25568151473999, "target_z": 1.9843749428969448, "ce": 6.1894097328186035, "genuineness": 3.5345609188079834, "identity": 0.3311350643634796, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/bbf0a8163773f5ebd39415eec912fd7cfa313c19fec044407cbdd5a30b30f801.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "bc95578b1f07d33e62fd1c8bffbbd89f09ab24336a5431667bc917dc774c0cec", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "16d14d27dc299b97581b6d460a57c939ad3c61c09557443400d907d4bf8eb478", "generation_seconds_batch": 23.995674683945253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.795957326889038, "target_z": 1.6640624351101645, "ce": 6.117925643920898, "genuineness": 4.005263805389404, "identity": 0.2939111292362213, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/bc95578b1f07d33e62fd1c8bffbbd89f09ab24336a5431667bc917dc774c0cec.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "bc9ebe851701dd4bc52632c1bcbad75b6aba5fb3f98480fd3adce7cc4853c33f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "80677a8e6228e3b61ce3baa37e6d09f620305c1a722b502ab26f376596f53981", "generation_seconds_batch": 17.14099936396815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2465310096740723, "target_z": 1.281250031147121, "ce": 6.117925643920898, "genuineness": 3.7542223930358887, "identity": 0.31592172384262085, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/bc9ebe851701dd4bc52632c1bcbad75b6aba5fb3f98480fd3adce7cc4853c33f.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "bcd18dd063352207b87cf161c9961abb182aeef603e9847a02628514c6f166bd", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "eac22495ebbfb2d14c8b72524b575dc285f23d207de79bd6757cf5833265b1aa", "generation_seconds_batch": 25.622621134039946, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25568151473999, "target_z": 1.9843749428969448, "ce": 6.23706579208374, "genuineness": 3.512594699859619, "identity": 0.17617857456207275, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinfrauen.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/bcd18dd063352207b87cf161c9961abb182aeef603e9847a02628514c6f166bd.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "bcf5568f9d4b140f1d9f3acbb2af18c4e9af02cf8d326b3cf3be2e61fb45b269", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b0b328ec1716e26651b422ad612c9b67e24aa5df1deb716d676803ca362e90a0", "generation_seconds_batch": 24.47086635499727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.569639682769775, "target_z": 2.2031250778678024, "ce": 6.348263263702393, "genuineness": 3.8797430992126465, "identity": 0.5478624105453491, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/bcf5568f9d4b140f1d9f3acbb2af18c4e9af02cf8d326b3cf3be2e61fb45b269.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": "bd02d35607c0f461a11eab3eaae1f94a3370551e68f84b8f19a028155ea76d43", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2847cc5e7887c6c7dd18c4e260ead69bab91112dd12f0b7685ceb0fe04b8b1c9", "generation_seconds_batch": 20.733893332071602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.479937553405762, "target_z": 2.1406251816915396, "ce": 5.967014789581299, "genuineness": 4.218649387359619, "identity": 0.5366578102111816, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/bd02d35607c0f461a11eab3eaae1f94a3370551e68f84b8f19a028155ea76d43.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "bd7bc614f8fedb3134814141ce4619e5186b1a2f3db1e67a99d68b340e0e6652", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "4d00e4862da904090287cae214d6c4e798698d68c8e042bae3e86a6027fde493", "generation_seconds_batch": 16.722289137076586, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5604889392852783, "target_z": 1.5, "ce": 6.078212261199951, "genuineness": 3.8797430992126465, "identity": 0.35625699162483215, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/bd7bc614f8fedb3134814141ce4619e5186b1a2f3db1e67a99d68b340e0e6652.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "bd8a33c7284f13089b4988d6c221e4cdbfc3438b1de3f712c5792e921d85ee70", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d500d8a5b9aedd942bb79bf6d911f55c29b55e725811e237bafd75ab81302da", "generation_seconds_batch": 24.27685255196411, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.278107166290283, "target_z": 2.0, "ce": 6.165581703186035, "genuineness": 4.218649387359619, "identity": 0.7211666703224182, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/bd8a33c7284f13089b4988d6c221e4cdbfc3438b1de3f712c5792e921d85ee70.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "bdcb5a5e9bdbf0ec3c78ebf061799866d518bae8e703cbd1a9d406477a16d2bc", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "15a9df1d11105adcb7260bffae536c9296f17541fb90656ffbfefcdce0ca18fd", "generation_seconds_batch": 16.12572433904279, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.717468023300171, "target_z": 1.609375067485429, "ce": 6.316492557525635, "genuineness": 4.281409740447998, "identity": 0.3792989253997803, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/bdcb5a5e9bdbf0ec3c78ebf061799866d518bae8e703cbd1a9d406477a16d2bc.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "bdd2a3a1bbcc859e5151dde7ab03b7a8be0ddf6fd1f293855eba5de27e82f488", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "7d7623a65378f432200d89478ff30c7d056e7526836c806da2b4b90ed61d016c", "generation_seconds_batch": 16.677778678014874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8652963638305664, "target_z": 1.0156250571030552, "ce": 6.173524379730225, "genuineness": 4.15588903427124, "identity": 0.6560534238815308, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/bdd2a3a1bbcc859e5151dde7ab03b7a8be0ddf6fd1f293855eba5de27e82f488.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "bea67fc3e096293040cdb4626b9f6b609d5580fdb2b9075fdedd0afd10162188", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "b2def2cf479a296ce3df4201e6421538a46b27faea0f7f4e49a4d1f10fc46528", "generation_seconds_batch": 17.786695753922686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.109915733337402, "target_z": 1.8828127361990015, "ce": 6.284721851348877, "genuineness": 4.13078498840332, "identity": 0.40055036544799805, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/bea67fc3e096293040cdb4626b9f6b609d5580fdb2b9075fdedd0afd10162188.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "bedeed83bfc327b487f5d870fef1c1b2e5a098abf315402c9f690dbb64a15f11", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "fbe64e96edf70d9d4d260203d697611bec54c076475cc415ec05731de8eddfa4", "generation_seconds_batch": 24.465058577014133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.695042371749878, "target_z": 1.5937500103823736, "ce": 6.23706579208374, "genuineness": 3.9048471450805664, "identity": 0.1907258927822113, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/bedeed83bfc327b487f5d870fef1c1b2e5a098abf315402c9f690dbb64a15f11.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "bf2899f98885ea55ae4ef2f813c3e492367b96ef4291453ffa1f92f70f8a243b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "a2ba03af352a0e1ea484374e1b8d07ab69f92773cd6331755dee552b03f4ed44", "generation_seconds_batch": 22.983637726982124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2465310096740723, "target_z": 1.281250031147121, "ce": 6.165581703186035, "genuineness": 3.936227321624756, "identity": 0.4686583876609802, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/bf2899f98885ea55ae4ef2f813c3e492367b96ef4291453ffa1f92f70f8a243b.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "bfd0e466843d943831bf5dd5032f21d83b393e239fe1776f26f48ba8ad336fc0", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "a9b3963bf7fc53667e9b1f3587fc40065a894f6e500197efe49b2a6b347d7729", "generation_seconds_batch": 23.404025606927462, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.381084442138672, "target_z": 1.3750000415294947, "ce": 5.982900142669678, "genuineness": 3.9676074981689453, "identity": 0.25748103857040405, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/bfd0e466843d943831bf5dd5032f21d83b393e239fe1776f26f48ba8ad336fc0.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": "c07084e693163e6601e61b718113370ec0aedb24d04f32c7d95ff93d2bb8e692", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f794023cfa26b682c641ac9d19fe2a162c1eb0a526756ad7fcb52e58895a5a1", "generation_seconds_batch": 18.065647045965306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5247883796691895, "target_z": 2.171874963661692, "ce": 6.09409761428833, "genuineness": 4.400654315948486, "identity": 0.7335807085037231, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, lis le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c07084e693163e6601e61b718113370ec0aedb24d04f32c7d95ff93d2bb8e692.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "c09be268178e3ebbfae9b749e961629a8b3148844e246e9de9c295fe7af63bad", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "cf6efee38f2f8b7e2fa9a4810d2955c510a2cf5548aa4e97aa4a49016bf6dcb2", "generation_seconds_batch": 15.311973260017112, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.143553733825684, "target_z": 1.9062499896176264, "ce": 6.0305562019348145, "genuineness": 3.6067352294921875, "identity": 0.2536791265010834, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinfliegen.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c09be268178e3ebbfae9b749e961629a8b3148844e246e9de9c295fe7af63bad.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "c0ff409a87fc2ecbb80760ba438de996ed527211270125cc2a8af07f7d93e1f6", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "27812c416563b6f1da9394942bf09df82ecd2ab5359aa359208dd7165c91bb97", "generation_seconds_batch": 23.87991921696812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.504425048828125, "target_z": 1.4609375233603408, "ce": 6.348263263702393, "genuineness": 4.582659721374512, "identity": 0.6393911838531494, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c0ff409a87fc2ecbb80760ba438de996ed527211270125cc2a8af07f7d93e1f6.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "c14ffd1c33b1773d5a76b59f54665d5bf8d17e2f1cfc80f12a9aed7d5701cd20", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "73e4803de37bb977db0adfe876c3724a810f5b1a22edbb61fa87b907e5c79a76", "generation_seconds_batch": 24.25161640997976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.121128082275391, "target_z": 1.890624932514571, "ce": 6.2767791748046875, "genuineness": 3.911123275756836, "identity": 0.3413267135620117, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c14ffd1c33b1773d5a76b59f54665d5bf8d17e2f1cfc80f12a9aed7d5701cd20.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "c1985b4378933c9ce82464978d19c5330bd4fafe68fb6d954955cf5b1879cec8", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7ba66826d08c487f6047208b4874196f10b0457f419ebb48622360e30534b7c1", "generation_seconds_batch": 23.970778897986747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7511062622070312, "target_z": 1.6328124870220329, "ce": 6.038498878479004, "genuineness": 4.042920112609863, "identity": 0.7169302701950073, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c1985b4378933c9ce82464978d19c5330bd4fafe68fb6d954955cf5b1879cec8.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": "c1a7d71867ef3d5e1517a7488dfea7ceb0274f1974ca948932da35f73a9b73d3", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "550344b0b8bb6c80c01d7f9e93f2fdce26d1162d024467a82203c6d2f6c1843c", "generation_seconds_batch": 26.433920657029375, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.098702430725098, "target_z": 1.8749998754115158, "ce": 6.300607204437256, "genuineness": 4.331617832183838, "identity": 0.6735574007034302, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade daneben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c1a7d71867ef3d5e1517a7488dfea7ceb0274f1974ca948932da35f73a9b73d3.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "c1ddd1d4344223d0207958d231fa0ce758253cc95c397c635fde66f7b0d4ec4f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "47ebd8a9febfee739379648b1cfda7c98557a1b309e2e069bca3c21fc019e556", "generation_seconds_batch": 28.71340654406231, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8071701526641846, "target_z": 1.671874963661692, "ce": 6.205295085906982, "genuineness": 3.9173994064331055, "identity": 0.27502772212028503, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c1ddd1d4344223d0207958d231fa0ce758253cc95c397c635fde66f7b0d4ec4f.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "c343ba7a40f4dbe7791d0b9aecad87dd9145f44b71f07ee1a791d4d6e9d4ad84", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "1fb867eebc8249bb8a8c1095104728d883f95854ebc04ea4023a1cf6baabba24", "generation_seconds_batch": 24.261287345085293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8071701526641846, "target_z": 1.671874963661692, "ce": 6.125868320465088, "genuineness": 4.2374773025512695, "identity": 0.6383188962936401, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c343ba7a40f4dbe7791d0b9aecad87dd9145f44b71f07ee1a791d4d6e9d4ad84.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9e17857d953f6abed965ee7b04bc473d1fa4078619e43f6ebefe52c6e6438397", "generation_seconds_batch": 17.31603498896584, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8968725204467773, "target_z": 1.7343750259559343, "ce": 6.062326908111572, "genuineness": 4.337893962860107, "identity": 0.7804495096206665, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "c3f1ceae1595eb67675b0b31492ca3f8d3a73b476170c1880bbed0045a251f03", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "cc1d5d6599e3c2e840eb1f3cb9794e23bf1ce2df747393a22d98da63db471a95", "generation_seconds_batch": 15.23614059900865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.818382978439331, "target_z": 1.6796874922132197, "ce": 6.213237762451172, "genuineness": 4.381826400756836, "identity": 0.5076826810836792, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c3f1ceae1595eb67675b0b31492ca3f8d3a73b476170c1880bbed0045a251f03.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": "c463df82f6e26a41987afa23931283b49536a1dc5323abf1ba1e248b4b9d456d", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3f008c47943c762f5a28b55ae8fdd11a359d55586ed6b8525215d851be800c37", "generation_seconds_batch": 20.663293187972158, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.457511901855469, "target_z": 2.1250001245884844, "ce": 6.332377910614014, "genuineness": 4.162164688110352, "identity": 0.6987568140029907, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c463df82f6e26a41987afa23931283b49536a1dc5323abf1ba1e248b4b9d456d.mp3" }, { "id": "heldout-nl-3", "language": "nl", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-3", "key": "c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f5128106d7b01fbbcd3e870a9fc32e17f8c58a048e57bf7ece6150af83cde6b4", "generation_seconds_batch": 18.55269476689864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1568286418914795, "target_z": 1.218749968852879, "ce": 6.23706579208374, "genuineness": 4.494794845581055, "identity": 0.8197141885757446, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee.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": "c4ab51c9531ecd3a130775a71292bd3f8f996e7512fdaf4abea339152b863d0e", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b7755fbbcafba743ebb2f22b9295a64464506504e682b46f22e57dca36d457c8", "generation_seconds_batch": 25.09835938096512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.840808629989624, "target_z": 1.695312549316275, "ce": 6.260893821716309, "genuineness": 4.049196243286133, "identity": 0.43053650856018066, "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_S_NARR_reg/c4ab51c9531ecd3a130775a71292bd3f8f996e7512fdaf4abea339152b863d0e.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "c5f87e4a8e025c3f98fa6cfdbef1ed0a112efe66c3598d9a4c9f8a3cf4d16511", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "7bc47f5d3a13f1b1819c91adb3c2f6915d75f4dffab80b2b0cb219a76196ee3c", "generation_seconds_batch": 23.069037634995766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3922972679138184, "target_z": 1.3828125700810223, "ce": 5.967014789581299, "genuineness": 3.747946262359619, "identity": 0.20348313450813293, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c5f87e4a8e025c3f98fa6cfdbef1ed0a112efe66c3598d9a4c9f8a3cf4d16511.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "c6cc186d724d8d1404d432ece1feacf9f4c9908e28de8fab90c54f93601aa94a", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "dd6be2c59655110338424ed65106b39e4c0c7994b447774aadc45d45eadc52e1", "generation_seconds_batch": 21.01452677801717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5380632877349854, "target_z": 1.4843749428969448, "ce": 6.117925643920898, "genuineness": 4.055472373962402, "identity": 0.42834022641181946, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c6cc186d724d8d1404d432ece1feacf9f4c9908e28de8fab90c54f93601aa94a.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "c74afb2826fe55b0bc8ba88711a69f1efe91428341ec9717f2d123c1b73bf599", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "29693cf60ff4b2d6daea0ee98aadf75e228c856a0db06b823dc8d27834191d8b", "generation_seconds_batch": 23.94573297793977, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.636916160583496, "target_z": 2.2499999169410105, "ce": 5.990842819213867, "genuineness": 3.9801597595214844, "identity": 0.5007643699645996, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c74afb2826fe55b0bc8ba88711a69f1efe91428341ec9717f2d123c1b73bf599.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "c76748173fba5228c217e55b9a25bed69163104cf80c5cef141896f66e374ad7", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "a3f084f9414bc80cbcc60369aaed8c52d0cb02de0e298a0872ea80bcf81bb7de", "generation_seconds_batch": 25.312129957950674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2108306884765625, "target_z": 1.9531251609267921, "ce": 6.221180438995361, "genuineness": 4.65797233581543, "identity": 0.5223773717880249, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c76748173fba5228c217e55b9a25bed69163104cf80c5cef141896f66e374ad7.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": "c7c875fdd8affd26736418066a63547d407a815259e2c1b6fe76c9696fa9ca13", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4e73c2f448618e2c8dbe602ebd1129ca48413bb5ec4ce56b92d2bb0ed2e347e7", "generation_seconds_batch": 18.218874246929772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7847447395324707, "target_z": 1.6562500726766158, "ce": 6.229123115539551, "genuineness": 4.419482707977295, "identity": 0.42612317204475403, "hypothesis": "De sleutels liggen op de keukentafel. Ik kijk in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c7c875fdd8affd26736418066a63547d407a815259e2c1b6fe76c9696fa9ca13.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "c7e47e73f3f7e672a5e48a7ad1dc9451c6857851c6658ef737c915382d80f798", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a6f5dbeaf58cac1e498f7eb3ca9beda72c5f8e0c4b2c86cad3bc9dc2bfb26f0c", "generation_seconds_batch": 15.91696778696496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8520214557647705, "target_z": 1.7031250778678027, "ce": 6.062326908111572, "genuineness": 4.055472373962402, "identity": 0.5955143570899963, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c7e47e73f3f7e672a5e48a7ad1dc9451c6857851c6658ef737c915382d80f798.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "c80a87c403fd0d7c955f0df3380d649750f7e9805ea47c7d373b5ce61faa743c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "1399a6a85d85d09bf4f13d91d9cc44e266212346bcdbc60c640845fbd5854016", "generation_seconds_batch": 20.45653994102031, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165979385375977, "target_z": 1.9218750467206815, "ce": 5.927301406860352, "genuineness": 4.394378662109375, "identity": 0.7027646899223328, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c80a87c403fd0d7c955f0df3380d649750f7e9805ea47c7d373b5ce61faa743c.mp3" }, { "id": "heldout-nl-0", "language": "nl", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-0", "key": "c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e998a9884afdd7974f642be077bb26182bcaebae82ec824bbb10528aeda5f4ae", "generation_seconds_batch": 17.073086285032332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.685891628265381, "target_z": 0.8906249325145711, "ce": 6.221180438995361, "genuineness": 4.519899368286133, "identity": 0.7549411058425903, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5a52bbb5c8263bf8faa7798cccf7572d67dc2bb5c9eba497c0f95b051d1ba2da", "generation_seconds_batch": 15.420115098007955, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3922972679138184, "target_z": 1.3828125700810223, "ce": 6.403861999511719, "genuineness": 4.469690799713135, "identity": 0.23681366443634033, "hypothesis": "Someone moved the little blue chair. It used to stand here beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "c8e41f970d2929f07410d920663829fa8f8cf69355da3703744264f604bd86e2", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "5b6dedadf0ed9c7c230f4181948716053c03e3f0d0c0badeeaaeeaf2b9797d7e", "generation_seconds_batch": 24.941238360013813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.177192211151123, "target_z": 1.9296875752722091, "ce": 6.157639026641846, "genuineness": 4.055472373962402, "identity": 0.7764143943786621, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c8e41f970d2929f07410d920663829fa8f8cf69355da3703744264f604bd86e2.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "c8f61038f251a71031df9f6940a970aef2917048266725374cf404d1f2fa22cf", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6d420737f438812b76014bbf8cf24b7c7ea0a6fc19ee2f043798389daf74abc3", "generation_seconds_batch": 24.54476727102883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.132340908050537, "target_z": 1.8984374610660988, "ce": 6.213237762451172, "genuineness": 4.356721878051758, "identity": 0.7768546938896179, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/c8f61038f251a71031df9f6940a970aef2917048266725374cf404d1f2fa22cf.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "cadda279a4abb1b63c62f9db2a20e51415cd02753a43d483ebec9228709b036b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "c85499158e99121467569d85c5d636d08596a8ff4a0c84fe916e9afe4df011c8", "generation_seconds_batch": 23.183536314056255, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2332563400268555, "target_z": 1.9687502180298475, "ce": 6.348263263702393, "genuineness": 4.099404335021973, "identity": 0.378798246383667, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/cadda279a4abb1b63c62f9db2a20e51415cd02753a43d483ebec9228709b036b.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "cb7b6629ca313f0dd9280c0e7e9b80ce4e842644a9d42ab32a0fed5a08379fbd", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "83546e64ad5ba70b311c2ff05d53362daac56a178cfe0863206dc94c70b61ef9", "generation_seconds_batch": 18.49656447500456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.941723585128784, "target_z": 1.7656249740440657, "ce": 6.109982967376709, "genuineness": 4.507347106933594, "identity": 0.5757132172584534, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/cb7b6629ca313f0dd9280c0e7e9b80ce4e842644a9d42ab32a0fed5a08379fbd.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "cbdd6c7d4c1b4b625f048188593b3f52618a49381d38c781af19ef1dba7e94e1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "474c3c0c3e07692d6ad9bebe96b6c149c07e754d13d2e00f1027ebcd808e411c", "generation_seconds_batch": 15.957069286028855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2241053581237793, "target_z": 1.2656249740440657, "ce": 6.125868320465088, "genuineness": 3.8671910762786865, "identity": 0.4376911520957947, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/cbdd6c7d4c1b4b625f048188593b3f52618a49381d38c781af19ef1dba7e94e1.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "cc1c53e6157ea8258cbea60222f057a28e00c569e445d27d55f2b0ab59d6321b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fad3df9111d7290a7c7ce235ca95ba96de6ad544f1799b38798bf144cd709005", "generation_seconds_batch": 22.18837694893591, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3474459648132324, "target_z": 1.3515624558749118, "ce": 5.967014789581299, "genuineness": 3.537698984146118, "identity": 0.4897395670413971, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/cc1c53e6157ea8258cbea60222f057a28e00c569e445d27d55f2b0ab59d6321b.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "cc29c26fc99ecfb87b8cd1205304547a24c0e5df61772b85201456278960cd36", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "9692e3d4a4714f9a558989fd63188c33527c95ff97021315b3bb9a8f9170ae2f", "generation_seconds_batch": 23.92989211890381, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5156378746032715, "target_z": 1.4687500519118684, "ce": 6.1894097328186035, "genuineness": 3.8922953605651855, "identity": -0.06889401376247406, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/cc29c26fc99ecfb87b8cd1205304547a24c0e5df61772b85201456278960cd36.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "cc72f45fbf222aad4c4a14f1721b41bf1cce04422b92e37afff23ae887f22aff", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ddd58e2a01a41c4785253db2385466b46d454ec8063762a86a49ff974de26cd0", "generation_seconds_batch": 22.75590888503939, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.336233139038086, "target_z": 1.3437499273233842, "ce": 6.117925643920898, "genuineness": 3.7918784618377686, "identity": 0.39466506242752075, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/cc72f45fbf222aad4c4a14f1721b41bf1cce04422b92e37afff23ae887f22aff.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "cc7d2ac0caee897bad5a4ce656b5eb08d8ea6fea1419b614d7c733047b424290", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "14aae78903552837d24a300d45355e74d2adc53380b939411f919b042d5aacd3", "generation_seconds_batch": 18.01036707102321, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8877220153808594, "target_z": 1.0312501142061106, "ce": 6.3641486167907715, "genuineness": 4.218649387359619, "identity": 0.20124095678329468, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/cc7d2ac0caee897bad5a4ce656b5eb08d8ea6fea1419b614d7c733047b424290.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "cc7e50652529ce9ce60510c81a8d52d519f2b8b19cded73f9a5df6c5e1dfa8f1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "329cea07ae79ade460d87a5adf26f8f05508186aea8ca4b1b8dca85366417546", "generation_seconds_batch": 22.75590888503939, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.650191307067871, "target_z": 1.5625000622942422, "ce": 6.1894097328186035, "genuineness": 3.71028995513916, "identity": 0.3282637596130371, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/cc7e50652529ce9ce60510c81a8d52d519f2b8b19cded73f9a5df6c5e1dfa8f1.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": "cd1f48bca8d14378ab75d8f1b5c0f9e767163dcef7a401bd197e5aaa899148d9", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The road turns left behind the bridge.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6602532d7e6abdafeba341fcd90c0d556e7b79fa51a5fbe0442b1cef4ca6c25b", "generation_seconds_batch": 22.960852851043455, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.300532817840576, "target_z": 2.015625057103055, "ce": 6.38797664642334, "genuineness": 3.936227321624756, "identity": 0.3208416998386383, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/cd1f48bca8d14378ab75d8f1b5c0f9e767163dcef7a401bd197e5aaa899148d9.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "cd88fba48f9b3321354fd51f48d6ba8664dc6357910de4180970801f085a578c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "108ff9afff7fd3512d0ca7adc904c005f877ad14679498ee0aa8f592d1f7b220", "generation_seconds_batch": 23.06795462197624, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.964149236679077, "target_z": 1.781250031147121, "ce": 6.268836498260498, "genuineness": 3.886019229888916, "identity": 0.3779059052467346, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/cd88fba48f9b3321354fd51f48d6ba8664dc6357910de4180970801f085a578c.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": "cd91666ae6f8a4f040ae74245134008ff7f7d066c274bad8c001e3a6bc6e01e3", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c83a720f9160d0b07e49f8571ffbf8c5335b067ffea3cb455b4562d3dfa3f94c", "generation_seconds_batch": 15.633880621986464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.704193115234375, "target_z": 2.2968750882501765, "ce": 6.340320587158203, "genuineness": 4.432034492492676, "identity": 0.5966012477874756, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón, y luego, junto a la ventana, llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/cd91666ae6f8a4f040ae74245134008ff7f7d066c274bad8c001e3a6bc6e01e3.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "ce53026874c4c5326a40c1c02e97dcd91fa3b3739d1c736ec0b0139e0aad4441", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "ace85d66f97bdf100ba7ac94388f993ee5ff1c7c69a264d4bae81257875c8360", "generation_seconds_batch": 23.87991921696812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.941723585128784, "target_z": 1.7656249740440657, "ce": 6.0146708488464355, "genuineness": 4.356721878051758, "identity": 0.6584562063217163, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ce53026874c4c5326a40c1c02e97dcd91fa3b3739d1c736ec0b0139e0aad4441.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "cf52eb0c228e2072d484bc25589caf23e0a150bfde0a6359ff538589987470e3", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "63e1a982ace183118267f7bdf5fefc318df1e161f60e1618a141efb51841b014", "generation_seconds_batch": 17.61245301202871, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8295958042144775, "target_z": 1.6875000207647475, "ce": 6.09409761428833, "genuineness": 4.143336772918701, "identity": 0.7250997424125671, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/cf52eb0c228e2072d484bc25589caf23e0a150bfde0a6359ff538589987470e3.mp3" }, { "id": "heldout-de-3", "language": "de", "sentences": [ "Jemand hat den kleinen blauen Stuhl umgestellt.", "Er stand hier... neben dem hohen Bücherregal.", "Wir stellen ihn nach dem Essen zurück." ], "text": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier... neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-3", "key": "cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d3d0c05ae72a008fe17adbf465f72ddb414ffe80ec86cdef1e396a81ed0f726e", "generation_seconds_batch": 18.070300854975358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2108306884765625, "target_z": 1.9531251609267921, "ce": 6.356205940246582, "genuineness": 4.300237655639648, "identity": 0.4980347156524658, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "cfe7f2a21404ff241430f1738f20949e93e2ba645693f01ba64836d15e56eb57", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "d611bfd5561d5256723083362d585bea759900ef07a5bcdc755a3e43457c1f3f", "generation_seconds_batch": 23.550526696955785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.479937553405762, "target_z": 2.1406251816915396, "ce": 6.1894097328186035, "genuineness": 4.218649387359619, "identity": 0.6041475534439087, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/cfe7f2a21404ff241430f1738f20949e93e2ba645693f01ba64836d15e56eb57.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "d0005a7ed7641a207a7d7d3ff1c1d737cb6d60d11245062843792f1031383698", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b8571e279111a0e6f5edc56b734bc0cec8077e69c3c8a8424206b3df7297c9a3", "generation_seconds_batch": 23.982853319961578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7847447395324707, "target_z": 1.6562500726766158, "ce": 6.213237762451172, "genuineness": 3.9425034523010254, "identity": 0.6852644681930542, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d0005a7ed7641a207a7d7d3ff1c1d737cb6d60d11245062843792f1031383698.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "d0c038ed59ca788b4f2f7967ad241d4e328981c29fa9ad7d5a264173c81ba42f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "32ffc935d4d59e868c754afc2848a7f94d4021d6b70f6ed0b288920f55d8951b", "generation_seconds_batch": 17.182650041067973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9101474285125732, "target_z": 1.0468750051911868, "ce": 5.95112943649292, "genuineness": 3.886019229888916, "identity": 0.6734412312507629, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d0c038ed59ca788b4f2f7967ad241d4e328981c29fa9ad7d5a264173c81ba42f.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "d0c416b482d10f6b2b43d84d0ec77993f17e8526677f486688baeed2ace8c670", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "8b08dd82de7aedbec0f06b9b49e045c495d1d8c830b1da50f0481db90c384811", "generation_seconds_batch": 24.461215593968518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.549276113510132, "target_z": 1.4921874714484724, "ce": 5.967014789581299, "genuineness": 3.986435890197754, "identity": 0.30815625190734863, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d0c416b482d10f6b2b43d84d0ec77993f17e8526677f486688baeed2ace8c670.mp3" }, { "id": "heldout-en-4", "language": "en", "sentences": [ "I wrote the number on the envelope.", "Read it carefully... and check the last two digits.", "Call me when you have a moment." ], "text": "I wrote the number on the envelope. Read it carefully... and check the last two digits. Call me when you have a moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-4", "key": "d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6a71f6d833eb803a7917214863b6c809f3edf7c094fb1d1a7ddb30b28554377a", "generation_seconds_batch": 17.392608234891668, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.100764751434326, "target_z": 1.1796874922132197, "ce": 6.356205940246582, "genuineness": 4.20609712600708, "identity": 0.33316516876220703, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "d120a606f682ae30259e8d96e243b56bbddb36dab1c83d88313e27c1032e2069", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0e3ce6cb0d71d9ba25e3c11b347f111b6d2fd123c1b89857b2d1df508f7291ed", "generation_seconds_batch": 24.225607201922685, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.977424144744873, "target_z": 1.0937500103823736, "ce": 6.086154937744141, "genuineness": 3.986435890197754, "identity": 0.3307691514492035, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d120a606f682ae30259e8d96e243b56bbddb36dab1c83d88313e27c1032e2069.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "d157f90e8b47119a0b296074c0f6e1d6ef0fe60f85bbff2e093b64e1098fd597", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "36abb36bf6ab5ea821628dc592b2f3b7ac24cc63f26d90ed28f1dea6bafa7026", "generation_seconds_batch": 23.39317159296479, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8968725204467773, "target_z": 1.7343750259559343, "ce": 6.221180438995361, "genuineness": 4.344170093536377, "identity": 0.7907687425613403, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d157f90e8b47119a0b296074c0f6e1d6ef0fe60f85bbff2e093b64e1098fd597.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "d1769b7ac870f79c4d86d187de038729d5d1cdb54151feec0a037461798d9e12", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "948f4d9455e3936faccd716a6e43fad97ba8f1cc900bbcf914a1246394b28ff6", "generation_seconds_batch": 23.00775673496537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8968725204467773, "target_z": 1.7343750259559343, "ce": 6.300607204437256, "genuineness": 4.287685871124268, "identity": 0.5019280910491943, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d1769b7ac870f79c4d86d187de038729d5d1cdb54151feec0a037461798d9e12.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "d1949411ecc489d2c178c65de4d326ff71bf2c7a15570ce596627e0b8b14c7ba", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "01c46d0fc1f40a27f7f47750da13c0d2e7c0dbcd25c8bbcd8273e3e5b5f60f54", "generation_seconds_batch": 23.922300960985012, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.695042371749878, "target_z": 1.5937500103823736, "ce": 5.998785495758057, "genuineness": 3.7918784618377686, "identity": 0.1905449628829956, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d1949411ecc489d2c178c65de4d326ff71bf2c7a15570ce596627e0b8b14c7ba.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33bc4f0a2e05e3769d863c444b8730ce27d1695566cfbc7d442f921806ecd6aa", "generation_seconds_batch": 17.87595609703567, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.435086250305176, "target_z": 2.109375067485429, "ce": 6.229123115539551, "genuineness": 4.093128204345703, "identity": 0.6675928235054016, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "d27c51a994b0791ce343069fd57309ba8af06db0964aed4ac421e37b683744c7", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "86687378e1e6920d80e1c0c569a4b29402dc1bf262184a3b76c92e486837679b", "generation_seconds_batch": 22.238272781018168, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.605340003967285, "target_z": 1.5312499480881316, "ce": 5.863759994506836, "genuineness": 3.8420867919921875, "identity": 0.1659626066684723, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d27c51a994b0791ce343069fd57309ba8af06db0964aed4ac421e37b683744c7.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "d2818d4593adaa64c78adf98e678eb3745d7a883347524fd51fece76fb14211e", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8e90ad52dc988075b0fd29a0597657b095466612cb3c34497451ae17d9fd9605", "generation_seconds_batch": 18.10032077797223, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.636916160583496, "target_z": 2.2499999169410105, "ce": 6.300607204437256, "genuineness": 3.9048471450805664, "identity": 0.5514771938323975, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d2818d4593adaa64c78adf98e678eb3745d7a883347524fd51fece76fb14211e.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "d2c2d49f29e408c07552512e5ecfca248812311bbfa5e54bdb86cf07ea45fe5b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "1d3713d663fb8f9de3671f1afaad10136ec61cc545f19252120062d9c0546684", "generation_seconds_batch": 18.47922730492428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.15476655960083, "target_z": 1.914062518169154, "ce": 6.260893821716309, "genuineness": 3.9801597595214844, "identity": 0.3354730010032654, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d2c2d49f29e408c07552512e5ecfca248812311bbfa5e54bdb86cf07ea45fe5b.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "d35f88da9d1c702d985d3cc64aec621576a1a304925179598aa37fb96e1766f7", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "184af5b4a1ac8dbadbfd34563515238c5d22fb3622e3a819833be13737a596b6", "generation_seconds_batch": 24.644598889979534, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.861171722412109, "target_z": 2.4062498234996474, "ce": 6.23706579208374, "genuineness": 4.193545341491699, "identity": 0.5667752027511597, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d35f88da9d1c702d985d3cc64aec621576a1a304925179598aa37fb96e1766f7.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "d37bea7f69f8724affbf34b312b65df74cbd8db910caac574259cd2f3c711036", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "d1e9aa4b851bdcc282218544a3e315e8502126c77ddae0814d3b418da038056d", "generation_seconds_batch": 24.37193830590695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.143553733825684, "target_z": 1.9062499896176264, "ce": 6.23706579208374, "genuineness": 3.9927117824554443, "identity": 0.5333022475242615, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d37bea7f69f8724affbf34b312b65df74cbd8db910caac574259cd2f3c711036.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "d3eae0d900d26314413b9efdb1faaedd10af1a19a805f88ea9193aa454e98e95", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "048186b68e17d6684b6f0ba511c04683776daa18e97f37b482544c4fd611e127", "generation_seconds_batch": 17.53220678807702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.457511901855469, "target_z": 2.1250001245884844, "ce": 6.308549880981445, "genuineness": 4.061748504638672, "identity": 0.4690299332141876, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d3eae0d900d26314413b9efdb1faaedd10af1a19a805f88ea9193aa454e98e95.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "80faef1c796c65c1c45d7a2ddeb16f1ee8072f5c40d6660d5951ed2e055fa72f", "generation_seconds_batch": 17.49339515902102, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.479937553405762, "target_z": 2.1406251816915396, "ce": 6.141753673553467, "genuineness": 4.457139015197754, "identity": 0.7963573932647705, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "d43ab93a0a87a20dadf35dc64097006e34b7694b984054827bb6399bda4cd34d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "94b6a90f7027cfee0a4658c4095d7e35453f05433db42c20a253e1b0d2d68c33", "generation_seconds_batch": 24.465058577014133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8071701526641846, "target_z": 1.671874963661692, "ce": 6.086154937744141, "genuineness": 3.8797430992126465, "identity": 0.2688492238521576, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d43ab93a0a87a20dadf35dc64097006e34b7694b984054827bb6399bda4cd34d.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "d43bb7e380c7d13465b3e23de37a12cae91822b5f85f575c2ffd15df19aed37e", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1769c626eb51418b765c0b2a84800a8befd175190d0638fc9ed8fa4c913cb2e4", "generation_seconds_batch": 23.018800194957294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.199617385864258, "target_z": 1.9453123001393064, "ce": 6.459460258483887, "genuineness": 4.20609712600708, "identity": 0.4427725672721863, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d43bb7e380c7d13465b3e23de37a12cae91822b5f85f575c2ffd15df19aed37e.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": "d44161e9b4a13f42459425914e1af7fd3ff6f1f58dc4f03f858aa8da49013499", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a10c2ea59f4af9833e4d431b30b0317812b038212885de9cfba62f5b388c0a3a", "generation_seconds_batch": 20.663293187972158, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.502363204956055, "target_z": 2.1562502387945948, "ce": 6.467403411865234, "genuineness": 4.143336772918701, "identity": 0.6669173240661621, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d44161e9b4a13f42459425914e1af7fd3ff6f1f58dc4f03f858aa8da49013499.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "d54983517e64c9b7e2482f67d97e755737e9a4286c40f2afdb25bc8f9b800a07", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "22d300b840796adcec29a6c0c182e9a7eb1e63922f971d433af31163b633fd50", "generation_seconds_batch": 24.065010763937607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.650191307067871, "target_z": 1.5625000622942422, "ce": 6.173524379730225, "genuineness": 3.7165660858154297, "identity": 0.2770646810531616, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d54983517e64c9b7e2482f67d97e755737e9a4286c40f2afdb25bc8f9b800a07.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": "d5cc2a9f4c1cc2f41fbbac5aab411340526450f80322691e554b069dab063689", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7456804c2ed9881fa507e9189509bee278ce7aca0dd5d7b9aecad362c1e55a0f", "generation_seconds_batch": 22.407566937967204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5247883796691895, "target_z": 2.171874963661692, "ce": 5.919358730316162, "genuineness": 3.9425034523010254, "identity": 0.5495619773864746, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d5cc2a9f4c1cc2f41fbbac5aab411340526450f80322691e554b069dab063689.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": "d5f681630ba82491e572a28a2c23c7e36b2ea62376cdfc0467d286e28e5f2564", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b90704e9a6f6aeba3456e785651d6fa207dcee194836950e114c6fbba6032f3", "generation_seconds_batch": 18.16453671292402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6614038944244385, "target_z": 1.5703124247277909, "ce": 6.1894097328186035, "genuineness": 4.21237325668335, "identity": 0.7123520374298096, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi, quand nous pourrons parler ?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d5f681630ba82491e572a28a2c23c7e36b2ea62376cdfc0467d286e28e5f2564.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": "d6407518b3572d933e46f1451dc16ef471b282f5624ce51d74f90618afcb0532", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f6e0e9be8831bab8fcfb83b50c8c955ac35dbb314db96aee5eeb325a6c286ce4", "generation_seconds_batch": 19.81677030492574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.222043037414551, "target_z": 1.9609373572423618, "ce": 6.38003396987915, "genuineness": 4.256305694580078, "identity": 0.5956645011901855, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d6407518b3572d933e46f1451dc16ef471b282f5624ce51d74f90618afcb0532.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "d68e371e8d750eba5a46b841eec64049a40904672e8a3a7d90c51c1ee9a45437", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "ca25caf195aea6146fa2d7bdcc037d24709bc1ea8314dcf7ef7b487835c49b7d", "generation_seconds_batch": 24.41286914900411, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.764381170272827, "target_z": 0.9453124662572855, "ce": 5.669164657592773, "genuineness": 3.6789097785949707, "identity": 0.42741858959198, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d68e371e8d750eba5a46b841eec64049a40904672e8a3a7d90c51c1ee9a45437.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": "d6b6214edf8689d8d45d75b2ed0cd13a7472d3ae9c2ed0c057081f77d8f050a7", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7feefe2bf366ae556a43df08e98deccc8285631a8c6181063609f1b45f405660", "generation_seconds_batch": 18.090373842976987, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.605340003967285, "target_z": 1.5312499480881316, "ce": 6.181467056274414, "genuineness": 4.042920112609863, "identity": 0.6901021003723145, "hypothesis": "La route tourne à gauche après le pont. Continue. Jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d6b6214edf8689d8d45d75b2ed0cd13a7472d3ae9c2ed0c057081f77d8f050a7.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "d70a49c34f09c7556137977a41599d5f7ec990037ce94de5ef3ed3ce4b120a84", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "a6df5f2bc2f3ada9352160a40b5a250f57c4b54ca941407f1b148dd4f22cacc6", "generation_seconds_batch": 19.41214348305948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.278107166290283, "target_z": 2.0, "ce": 6.062326908111572, "genuineness": 4.369274139404297, "identity": 0.770392656326294, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d70a49c34f09c7556137977a41599d5f7ec990037ce94de5ef3ed3ce4b120a84.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "d741ef49721857674a21fb4f4eeeb4dfabeba25f75d6f9b1f114ae222ec2420d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "049bee5042aa5dc6277531210dbb95b1d915b113d300fde2d56ca4ab6d6721d4", "generation_seconds_batch": 18.212289302959107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.863234043121338, "target_z": 1.7109374403013513, "ce": 6.0305562019348145, "genuineness": 3.9676074981689453, "identity": 0.6128065586090088, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d741ef49721857674a21fb4f4eeeb4dfabeba25f75d6f9b1f114ae222ec2420d.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "d7f454fadb81827bf8d8b9728fd87ae20d253774bf4c5460b8271d490fb632f8", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d2eb2926e1750a092fd307572d45c70b503055a188966002b677e7839c68f05c", "generation_seconds_batch": 23.915609329007566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9753620624542236, "target_z": 1.7890625596986487, "ce": 6.2767791748046875, "genuineness": 3.8922953605651855, "identity": 0.5203163623809814, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d7f454fadb81827bf8d8b9728fd87ae20d253774bf4c5460b8271d490fb632f8.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "d846845cfda644039d1beac10cfd4f0a2c4117513677dd1c04fd3f009f991480", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "4d5048ce772f2a50fddcfcf3a3bef51f83c764ba0e45f2cd81933e2db0aeb802", "generation_seconds_batch": 24.644598889979534, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.749044418334961, "target_z": 2.328125202456287, "ce": 6.117925643920898, "genuineness": 3.998987913131714, "identity": 0.6897903084754944, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d846845cfda644039d1beac10cfd4f0a2c4117513677dd1c04fd3f009f991480.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "d86e2e197064eaf84ef261c611fe3bbe72c4782df7e0c4881fc5a529ed462067", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e8b9dd8324f1f0a5bd971c2d2a7bb12be4627ad63bf4655bbe153c75163f5d2c", "generation_seconds_batch": 17.974390335031785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.181467056274414, "genuineness": 3.8671910762786865, "identity": 0.40514636039733887, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d86e2e197064eaf84ef261c611fe3bbe72c4782df7e0c4881fc5a529ed462067.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "d8c3c66e54e4dbc463b36785052160cbff69949e4831a4ad426e34df3cb98715", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "ba45ae8acc7c1cf08c440197f3af1f3788ac2e01aae152aba1f7fb79a2d96df9", "generation_seconds_batch": 24.327050105086528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.435086250305176, "target_z": 2.109375067485429, "ce": 6.197352409362793, "genuineness": 3.8044304847717285, "identity": 0.7025346755981445, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d8c3c66e54e4dbc463b36785052160cbff69949e4831a4ad426e34df3cb98715.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "d8ce5801cf2a956e3f565e11b24528e6ea367eb5cb3ea199def7ed4c8212dc34", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "39520bd142349ce25a1aa0eddf9406a5ca043af3cdaed224db5a6f654954f0b4", "generation_seconds_batch": 24.384073166991584, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2689566612243652, "target_z": 1.2968750882501763, "ce": 6.181467056274414, "genuineness": 3.8922953605651855, "identity": 0.27714210748672485, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d8ce5801cf2a956e3f565e11b24528e6ea367eb5cb3ea199def7ed4c8212dc34.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "d945298410a22d43f990d06b982ccdef53bf097d6531918e31a76fb9b2323824", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e9cc9fa97f9abc4385a53c53abe19872ec0f0e326a38822e2bcb6ce5cf13f44", "generation_seconds_batch": 24.364615836995654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5247883796691895, "target_z": 2.171874963661692, "ce": 6.086154937744141, "genuineness": 3.9299514293670654, "identity": 0.6005227565765381, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d945298410a22d43f990d06b982ccdef53bf097d6531918e31a76fb9b2323824.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "d9dbbf7621aef56e5786a0dc095ec023f7ec7d49c2617605dba38edd95f08489", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "efcc47ae2a06efe2a2a3dc19f3c799099269721d4b9b3eaee740a81cdde3b727", "generation_seconds_batch": 23.72428324096836, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.199617385864258, "target_z": 1.9453123001393064, "ce": 6.205295085906982, "genuineness": 4.293961524963379, "identity": 0.6828197836875916, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d9dbbf7621aef56e5786a0dc095ec023f7ec7d49c2617605dba38edd95f08489.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "d9ec672aa5a05aab84988a79d41fe9c358d3c2865706a18f467cda7bc0fd3ea7", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "40bdf5cd5c1e8d58b78f6dbb11d8507ac06cf5b361096a71daf00b9c9a9aa7a1", "generation_seconds_batch": 18.47922730492428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.717468023300171, "target_z": 1.609375067485429, "ce": 6.133810997009277, "genuineness": 4.149612903594971, "identity": 0.5162839889526367, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/d9ec672aa5a05aab84988a79d41fe9c358d3c2865706a18f467cda7bc0fd3ea7.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "da95efc1d029c954220166a7c19719f9d9e3443a5b4c8636478cab52572cb640", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "9d925e680ed88e3be64abc6202ad48cc051d226fcc64dbbcb84b53c91f7383f4", "generation_seconds_batch": 24.3537193860393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.659341812133789, "target_z": 2.2656249740440657, "ce": 5.982900142669678, "genuineness": 4.030367851257324, "identity": 0.5996068716049194, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/da95efc1d029c954220166a7c19719f9d9e3443a5b4c8636478cab52572cb640.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "db149bef931fddb8032e8c24aff1c4cd460203bbc60dc87e36d48581a2a30825", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "2078878e011419b74cd2bbf331a3e2f96d1182e525731d5609ad863a24ff600e", "generation_seconds_batch": 24.950757495942526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.569639682769775, "target_z": 2.2031250778678024, "ce": 6.054384231567383, "genuineness": 4.3127899169921875, "identity": 0.6312478184700012, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/db149bef931fddb8032e8c24aff1c4cd460203bbc60dc87e36d48581a2a30825.mp3" }, { "id": "heldout-en-0", "language": "en", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-0", "key": "db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "210acc4a39cafcbc6479f7ab3677186de7bde892d3e1064333d473f741cbf346", "generation_seconds_batch": 17.811453575035557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.044700860977173, "target_z": 1.1406250155735604, "ce": 6.419747352600098, "genuineness": 4.2374773025512695, "identity": 0.2119649350643158, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679.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": "dbd6f85eb794f849a56136472dd493988c00ff7d811df94944c2fda8f3533c88", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d99742c9d9b8418095f65a3793a209b5baf5eae36a23b7c3c7a15fbc37d869e6", "generation_seconds_batch": 18.02554661198519, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.717468023300171, "target_z": 1.609375067485429, "ce": 6.332377910614014, "genuineness": 4.2374773025512695, "identity": 0.7359306812286377, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/dbd6f85eb794f849a56136472dd493988c00ff7d811df94944c2fda8f3533c88.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "dbec019324a384c851b0baca57d7ab74648e7e08712f74af4f657d72bbcc840b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "f2f531bf81d6a2dd2b497e16c2a15a358146a80dcd071b3112936cd5d4486d7a", "generation_seconds_batch": 18.027226150035858, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5156378746032715, "target_z": 1.4687500519118684, "ce": 6.308549880981445, "genuineness": 4.469690799713135, "identity": 0.512279748916626, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/dbec019324a384c851b0baca57d7ab74648e7e08712f74af4f657d72bbcc840b.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "dc50c07e7a64589ccd0d6c84ddb55e72ab85f73df9743d40653d69c9377ed47e", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6eec200df0582892d14311e8d3c37d07345ca3b023b189c1ef8d114fa251f388", "generation_seconds_batch": 23.29165158199612, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.300532817840576, "target_z": 2.015625057103055, "ce": 6.205295085906982, "genuineness": 3.7793264389038086, "identity": 0.4022444486618042, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/dc50c07e7a64589ccd0d6c84ddb55e72ab85f73df9743d40653d69c9377ed47e.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "dc910db8c62d191e9377cd4e2885b68117f7e47591fafb14ad7d57b8ef29a144", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "4bc5e10ca0cd3d5274d2768aa2c02ded7ee5352a5aaf022e4873f2ae75b2d0dd", "generation_seconds_batch": 26.431296546012163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.941723585128784, "target_z": 1.7656249740440657, "ce": 5.704906940460205, "genuineness": 3.798154592514038, "identity": 0.4792335629463196, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/dc910db8c62d191e9377cd4e2885b68117f7e47591fafb14ad7d57b8ef29a144.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "dcd0ed3e6e77479f8291ddc8a7a2f7521e3124ef17f528f6549bc1e9f80bc01d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "6bb6ec0c08489299ed5915ac119e6e3e76931c963a4733bfa81fb2af3ba690a5", "generation_seconds_batch": 26.069628403987736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.39023494720459, "target_z": 2.0781249532793185, "ce": 6.372091293334961, "genuineness": 4.162164688110352, "identity": 0.40797317028045654, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/dcd0ed3e6e77479f8291ddc8a7a2f7521e3124ef17f528f6549bc1e9f80bc01d.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "dcddf525e12ecb8e6dace71238a420256b1c507ac4526958bdc2d0ed171b998f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "057457bf7b23ae3826670270a6e7b72a6db56f13e1fa6b42bd81093d3f068f44", "generation_seconds_batch": 23.00775673496537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8071701526641846, "target_z": 1.671874963661692, "ce": 6.435632705688477, "genuineness": 4.243753433227539, "identity": 0.4418720304965973, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/dcddf525e12ecb8e6dace71238a420256b1c507ac4526958bdc2d0ed171b998f.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "dd7e8fdc9a04039752031576222d2d15422ccd207d6bd3bc61cea2320142ceb9", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "8a97f055ec8e4ca5bfefd59eee893422a93d5de71d5c8d3a1225228c51b24463", "generation_seconds_batch": 22.561396892997436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7847447395324707, "target_z": 1.6562500726766158, "ce": 6.09409761428833, "genuineness": 4.055472373962402, "identity": 0.4659833312034607, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/dd7e8fdc9a04039752031576222d2d15422ccd207d6bd3bc61cea2320142ceb9.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "ddbf2171192607cdb6ee2a75f3fef2e267109cb13b4e0016300cf54207954dcc", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3683d7750a6bbcd75633bf61f5386c498966b0b0a2eade800b5a3318e137459b", "generation_seconds_batch": 26.767446773010306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.358658790588379, "target_z": 1.3593749844264396, "ce": 6.0146708488464355, "genuineness": 3.722842216491699, "identity": 0.2848331332206726, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ddbf2171192607cdb6ee2a75f3fef2e267109cb13b4e0016300cf54207954dcc.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "de7caec4015d1f2c4aebf3451a2282f5f7e4f7c5696bf9723875d08c392c7336", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "aba633ecbeaf9cc8b0f028f2ec9bc8588a0318dd42bfb4fd9003f7048f6070ce", "generation_seconds_batch": 23.494674212066457, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 6.0146708488464355, "genuineness": 3.936227321624756, "identity": 0.5563228130340576, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/de7caec4015d1f2c4aebf3451a2282f5f7e4f7c5696bf9723875d08c392c7336.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "de875a74b381d7ae3f2f7903230df75bce41db40f44b07ec5263a77f747c0d9f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d759e39f5310b4cd4d10ba3db00fed3060865f07c27a78a3ba71dfe5628c78f2", "generation_seconds_batch": 30.128549254965037, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.076276779174805, "target_z": 1.8593748183084604, "ce": 6.1894097328186035, "genuineness": 4.193545341491699, "identity": 0.565170407295227, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/de875a74b381d7ae3f2f7903230df75bce41db40f44b07ec5263a77f747c0d9f.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "dec5d8cf89893d942f19e5d6ff35445533070c052036cce5aba4ce4af4b3230a", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "1f45f00f69b6f6a095f5a90fb8be5f712d24c98ed822b546d21b351dcb4481d6", "generation_seconds_batch": 17.61831695807632, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2108306884765625, "target_z": 1.9531251609267921, "ce": 6.022613525390625, "genuineness": 4.337893962860107, "identity": 0.627091109752655, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/dec5d8cf89893d942f19e5d6ff35445533070c052036cce5aba4ce4af4b3230a.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "def93f7b340388bffa3a3fff87bf60c489527305302a15e1162580857490f80d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7dac8371bb4abc628c75b20b20ba89f1a31d8b691ed05834365f87ea07acfdea", "generation_seconds_batch": 19.363819874008186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3922972679138184, "target_z": 1.3828125700810223, "ce": 6.09409761428833, "genuineness": 3.7165660858154297, "identity": 0.10597331076860428, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/def93f7b340388bffa3a3fff87bf60c489527305302a15e1162580857490f80d.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "df1d55d9aab06f72828690b05dce09cd27b69e2a020e1821d19a40a0d38781a9", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "cd781eb5d26d7e9d01404b770685b30ca1a69c9cbb848361550d8c32ecea81f9", "generation_seconds_batch": 21.96367812389508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.504425048828125, "target_z": 1.4609375233603408, "ce": 6.205295085906982, "genuineness": 3.998987913131714, "identity": 0.6917428970336914, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Hein?", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/df1d55d9aab06f72828690b05dce09cd27b69e2a020e1821d19a40a0d38781a9.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": "e04f43f470bef4c7cf1ca8e63fceb6ed98f922d5330c2f21c14b57e2b727180e", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Your message arrived after I had left.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ebba4a8ce70d4d301f88c692209e0005048de3e49f15afd693d929805388ac1a", "generation_seconds_batch": 17.441698877955787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.571701765060425, "target_z": 1.5078125285515276, "ce": 6.419747352600098, "genuineness": 3.9173994064331055, "identity": 0.20509931445121765, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e04f43f470bef4c7cf1ca8e63fceb6ed98f922d5330c2f21c14b57e2b727180e.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "e095214b72492ef913459b1dfff8e62f0a2876cbc8470d51c6385b7d993f0a70", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "03929fae19a1c0dd83c0fe02f0a81f4b449935244736776d3de80726133f0029", "generation_seconds_batch": 22.363289171946235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.773531913757324, "target_z": 1.6484375441250882, "ce": 6.229123115539551, "genuineness": 4.15588903427124, "identity": 0.7110462784767151, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e095214b72492ef913459b1dfff8e62f0a2876cbc8470d51c6385b7d993f0a70.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "e0e45ec1cdaee7832fc281157bbf3d20efd2fa0b332a1d6261f4d0040692382f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5f1de7f3553dbb5968049be8a808d86f3dc62c87087d4915cdd774e431df6d7a", "generation_seconds_batch": 17.151270417962223, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25568151473999, "target_z": 1.9843749428969448, "ce": 6.165581703186035, "genuineness": 4.068024158477783, "identity": 0.5477871894836426, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e0e45ec1cdaee7832fc281157bbf3d20efd2fa0b332a1d6261f4d0040692382f.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "e12e45237e48c736011585018d273c3ad31fda13136a350e4eddc14ead66972a", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "5fb18b52be1d65eaf72119afc8d165b420b2443987c3486a2d0b7a7c096fd33f", "generation_seconds_batch": 18.434154095011763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.941723585128784, "target_z": 1.7656249740440657, "ce": 6.348263263702393, "genuineness": 4.432034492492676, "identity": 0.3563288450241089, "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_S_NARR_reg/e12e45237e48c736011585018d273c3ad31fda13136a350e4eddc14ead66972a.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "e1d557019d1922e7adf187070272b26935d2d6e6668d81bbcb2fea472dd76539", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "73b68b5ca373c5817211db2ad871d862e587fbb6528110c6461cad5657f53adc", "generation_seconds_batch": 24.127647768938914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.717468023300171, "target_z": 1.609375067485429, "ce": 6.0305562019348145, "genuineness": 3.8797430992126465, "identity": 0.23170645534992218, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e1d557019d1922e7adf187070272b26935d2d6e6668d81bbcb2fea472dd76539.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "e2a0c19118d583e1f60a95f4d18e52d50f820bb1f62ffaa614f7816c5e01db7d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "c7d11a56cf556164b7acdbc1b8f1f58d81f89faf814ac83f053f849259f7f924", "generation_seconds_batch": 23.72428324096836, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.086154937744141, "genuineness": 4.400654315948486, "identity": 0.6951457262039185, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e2a0c19118d583e1f60a95f4d18e52d50f820bb1f62ffaa614f7816c5e01db7d.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "e2ec8ae3385f008780ec33ddce3b387f7762c2ce0dd8488a84b06038e6d9a33f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af2db6a11061ce1f2d11c1e04e33ac3f760416e05df8956bae1ee4ea5b815a49", "generation_seconds_batch": 17.95437559799757, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.078339099884033, "target_z": 1.1640624351101645, "ce": 6.133810997009277, "genuineness": 4.250029563903809, "identity": 0.6028872728347778, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e2ec8ae3385f008780ec33ddce3b387f7762c2ce0dd8488a84b06038e6d9a33f.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "e2fd746fd397c69e695f8c393004aa74a7e2a6db3b931d8e95fe9b30002aad4d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "32bf85f50695ffd32f14d47c88588a44dced2aacf91d0c274db551b1b0451cbd", "generation_seconds_batch": 23.917659475002438, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7699875831604004, "target_z": 0.9492187305330494, "ce": 6.229123115539551, "genuineness": 4.344170093536377, "identity": 0.5669259428977966, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e2fd746fd397c69e695f8c393004aa74a7e2a6db3b931d8e95fe9b30002aad4d.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "e327640cf38911910a76229cf114d57cea4fadb759a608f30d09ee29d89bf926", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "e5b3fb83fcc9e96489ec562c2e9bbbfa704190c2e5c8c2a0e325e61a737606d5", "generation_seconds_batch": 18.644986761035398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9305107593536377, "target_z": 1.7578124454925381, "ce": 6.165581703186035, "genuineness": 3.9550557136535645, "identity": 0.6842095851898193, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e327640cf38911910a76229cf114d57cea4fadb759a608f30d09ee29d89bf926.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": "e37cf51200b4cb79e5835ac96ad32257b00fbd69c3a5dbc608e2fa794f2f3555", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Your message arrived after I had left.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "417f2a0ac0793b84a2e4e12f78a601aec116d7a82b4c3b70b78459c4b640590d", "generation_seconds_batch": 17.441698877955787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.85408353805542, "target_z": 1.0078125285515276, "ce": 6.292664527893066, "genuineness": 4.055472373962402, "identity": 0.35571086406707764, "hypothesis": "Your message arrived after I had left. I read it twice before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e37cf51200b4cb79e5835ac96ad32257b00fbd69c3a5dbc608e2fa794f2f3555.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "e3c46e44dc23f3d89e322f31c7112b435669a89d92c8d650fb2fd0261f97d7fc", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "eed8e6fa884e62f7e2011dce71d6dade0dd5b9d2aa891eefd0ea74d489a3df69", "generation_seconds_batch": 24.127647768938914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.695042371749878, "target_z": 1.5937500103823736, "ce": 6.0305562019348145, "genuineness": 3.936227321624756, "identity": 0.3846079707145691, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e3c46e44dc23f3d89e322f31c7112b435669a89d92c8d650fb2fd0261f97d7fc.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "e404b951f80a616d9683ab7ab804a5d7ff09fb3b29054e34ec827055fed42f6b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 129, "seconds_generated": 10.32, "mp3_sha256": "94cfcb2bbf458a534e46c5e5a003b0d4cc23348c5d1f8e57c9f6ad51b677181f", "generation_seconds_batch": 23.585260680993088, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.031425952911377, "target_z": 1.828125036338308, "ce": 6.0305562019348145, "genuineness": 4.174716949462891, "identity": 0.5120718479156494, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e404b951f80a616d9683ab7ab804a5d7ff09fb3b29054e34ec827055fed42f6b.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "e464cdf2ab249515ca2b0f6fa754d6ad41642537b50701b43bae5791f7d42d8c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "037cc5b1b4c522502634c464b2ed2737d21ae8cce9d97f182a6edf4f0ccdec13", "generation_seconds_batch": 25.659792028018273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7286806106567383, "target_z": 1.6171874299189777, "ce": 6.300607204437256, "genuineness": 4.268857479095459, "identity": 0.43284013867378235, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e464cdf2ab249515ca2b0f6fa754d6ad41642537b50701b43bae5791f7d42d8c.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "e49293dd84d9f8e099c4f44119596d5e4b441620593524e5eb6874c9ab2d4e20", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "b9d7755b89ef60666b1127f99baa9a49ba7c61ea39d571dbc5c8adcd4edc5a82", "generation_seconds_batch": 30.63822686392814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.547214031219482, "target_z": 2.1875000207647473, "ce": 5.919358730316162, "genuineness": 3.936227321624756, "identity": 0.6092978715896606, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e49293dd84d9f8e099c4f44119596d5e4b441620593524e5eb6874c9ab2d4e20.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": "e4e351dc212e6e95498d19153e5061407d8e010d0891aebc09cf9bd08d72a117", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "eee0be99ef738de53aa3c3a76a9493020673203d918e169bd390e5b3dbe08e3f", "generation_seconds_batch": 18.307272696984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.345383644104004, "target_z": 2.0468748390732077, "ce": 6.229123115539551, "genuineness": 3.9801597595214844, "identity": 0.513336718082428, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e4e351dc212e6e95498d19153e5061407d8e010d0891aebc09cf9bd08d72a117.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "e50659fb22508a7a5cd302e9a4405c0093cca199e5825ad8a5bfbfc086eed3bc", "bucket": "vn_S_NARR_reg", "phase": "wording", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "1a8f06596511186182e81252fd26281bf08cff88e9dd07ead8235b11b7d2f10e", "generation_seconds_batch": 18.078224043012597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.123190402984619, "target_z": 1.195312549316275, "ce": 5.895530700683594, "genuineness": 4.137060642242432, "identity": 0.5317518711090088, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e50659fb22508a7a5cd302e9a4405c0093cca199e5825ad8a5bfbfc086eed3bc.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "e5cb58f826879666a806bd3b5311404f813c069081fc817144fc6476e152cea1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "e757f05d1ae0ae8e410934e6de5891c6a3a6b1818ebc965ddc5bf66ab14b4488", "generation_seconds_batch": 24.0181316989474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25568151473999, "target_z": 1.9843749428969448, "ce": 6.213237762451172, "genuineness": 4.13078498840332, "identity": 0.5865187048912048, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e5cb58f826879666a806bd3b5311404f813c069081fc817144fc6476e152cea1.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "e5f3d18454e967a254058f167f0acee46802e9d098ee4bac5ed0dd771a9e99c0", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "e9fae0125b9dc94f23d5a09c8ee290971e666884acc4fd0cb806a90e1f22d093", "generation_seconds_batch": 23.2142091980204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.605340003967285, "target_z": 1.5312499480881316, "ce": 6.268836498260498, "genuineness": 4.268857479095459, "identity": 0.33748918771743774, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e5f3d18454e967a254058f167f0acee46802e9d098ee4bac5ed0dd771a9e99c0.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "e6134ae4f2a78cf3d511a03bcddf43343b2f1043ac5cc2dd10660c80b7b3e08d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c01e8e2d6d861ba9daa3e2e6cccc5122c9358d78435409b8366871bfd964d0cb", "generation_seconds_batch": 23.18086227006279, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.358658790588379, "target_z": 1.3593749844264396, "ce": 6.1894097328186035, "genuineness": 3.8797430992126465, "identity": 0.3261187672615051, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e6134ae4f2a78cf3d511a03bcddf43343b2f1043ac5cc2dd10660c80b7b3e08d.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e1bf8281a458f89fcb5d183a17950171c44a879baddca15f981b5f5aac0cd343", "generation_seconds_batch": 15.420115098007955, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9192981719970703, "target_z": 1.7500000830589895, "ce": 6.41180419921875, "genuineness": 4.337893962860107, "identity": 0.3084883689880371, "hypothesis": "Someone moved the little blue chair. It used to stand here beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "e728fdab30b70625ac7386f8a93cdb989510fa4494827ae2bbf85d03f9795b84", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bc22424325f4c2a280878bb4a97f84dd14c7f0bd899d86b722eecc3ac72753e9", "generation_seconds_batch": 24.54476727102883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9305107593536377, "target_z": 1.7578124454925381, "ce": 6.141753673553467, "genuineness": 4.218649387359619, "identity": 0.6579469442367554, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e728fdab30b70625ac7386f8a93cdb989510fa4494827ae2bbf85d03f9795b84.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "e7499fbc4876ad946e15696f479411637f71661cbd8af4c964f0290f52d12d0c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7cb5678b87521cf91d2954aa028093a1d1ba58d0dac8826fee807753ce23180a", "generation_seconds_batch": 24.274982999893837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.457511901855469, "target_z": 2.1250001245884844, "ce": 6.292664527893066, "genuineness": 3.7667744159698486, "identity": 0.2678833603858948, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e7499fbc4876ad946e15696f479411637f71661cbd8af4c964f0290f52d12d0c.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "e75412a477fdc02c4abd138152583f7ce8d07e0fd1a5c7a0265d060657f67011", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "a970d1f3ac335c7926879073ef350bce5518be30116ff8a1677a306bfaf4d393", "generation_seconds_batch": 23.861214244971052, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.098702430725098, "target_z": 1.8749998754115158, "ce": 6.197352409362793, "genuineness": 3.9299514293670654, "identity": 0.6740615963935852, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e75412a477fdc02c4abd138152583f7ce8d07e0fd1a5c7a0265d060657f67011.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": "e760f1c1a698c7a75a8a4169159dd61b345dc0f61bbeaf5ff66e81953ac110d1", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a6ece15d0392b36ff7b8ea5597e7bd937968fa9649675e9326da15aa36a13084", "generation_seconds_batch": 24.649570515961386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 6.1894097328186035, "genuineness": 3.8295347690582275, "identity": 0.6302297115325928, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e760f1c1a698c7a75a8a4169159dd61b345dc0f61bbeaf5ff66e81953ac110d1.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "e77ac0a286eca866bdd3faae41a34837f93e78fb2ed7738d08f40b2bc3377296", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "8438f6f8f7e321a6ee8ded81c159e1f479096a87ef1fef8f14de823004978ee6", "generation_seconds_batch": 24.895769884926267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.177192211151123, "target_z": 1.9296875752722091, "ce": 6.09409761428833, "genuineness": 4.093128204345703, "identity": 0.8284051418304443, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e77ac0a286eca866bdd3faae41a34837f93e78fb2ed7738d08f40b2bc3377296.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "e7936864f88507c9b198488ff30d4ee8098e709a969a035ffd035573fe4f2ce5", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "a6d9dbab04bae3ee537b72259ba928a763696dd49f2075064eea80fea2e20c93", "generation_seconds_batch": 29.00909856299404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.39023494720459, "target_z": 2.0781249532793185, "ce": 6.0146708488464355, "genuineness": 3.873466968536377, "identity": 0.3321041166782379, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e7936864f88507c9b198488ff30d4ee8098e709a969a035ffd035573fe4f2ce5.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "e7b2300fac2ae2a19e23ea4a7fdd8660740844bc58960006b730688e609ceaf9", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "d4e4cfa7ab24bd771dc3c80d0fdd53d8de595088d40a4d82d8c23d6742103b58", "generation_seconds_batch": 24.065010763937607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8071701526641846, "target_z": 1.671874963661692, "ce": 6.141753673553467, "genuineness": 3.7542223930358887, "identity": 0.25556614995002747, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e7b2300fac2ae2a19e23ea4a7fdd8660740844bc58960006b730688e609ceaf9.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": "e82f666a3cd9a5211ab02f110f7e1fb225881644dba97359b30bc141d916d70f", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Escribí el número en el sobre.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0fe7dba7562a1821d6e1d81c4146f3d343120e52176598f2a0825df85f96fddb", "generation_seconds_batch": 17.197586350957863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.121128082275391, "target_z": 1.890624932514571, "ce": 6.324435234069824, "genuineness": 4.356721878051758, "identity": 0.6780709028244019, "hypothesis": "Escribí el número en el sobre, léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e82f666a3cd9a5211ab02f110f7e1fb225881644dba97359b30bc141d916d70f.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "e858aaa2dad309f379f6e107b9e8e8d67d0ff3383fbc600570455541a901a1c4", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "5a8f2e164dfeac92023c1361cf15c5c9b81ea5d42a2b8807d21a1b4e934b1b4b", "generation_seconds_batch": 24.07227678806521, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1884050369262695, "target_z": 1.937500103823737, "ce": 6.1020402908325195, "genuineness": 3.7291181087493896, "identity": 0.5376412868499756, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e858aaa2dad309f379f6e107b9e8e8d67d0ff3383fbc600570455541a901a1c4.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "e85b60363fbcd098e3f646776a34bc6fd559062107916055a5ff94f0c3ea82cc", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52e74a7b55b61c0083052c435ebe029ae3170f8dc65791d066fb7f07ffbe7ef7", "generation_seconds_batch": 23.52070540492423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1568286418914795, "target_z": 1.218749968852879, "ce": 6.070269584655762, "genuineness": 3.8420867919921875, "identity": 0.3498755991458893, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e85b60363fbcd098e3f646776a34bc6fd559062107916055a5ff94f0c3ea82cc.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "e953ec125a570545657c651dde8f07adfbd24189372a388ccfe05be4c3e64e57", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "7cc995e70419f09eadb44abd5e2d0eed94346c4bf5915d505225236d1c2e894e", "generation_seconds_batch": 24.901143931085244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 5.927301406860352, "genuineness": 3.9550557136535645, "identity": 0.6040432453155518, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e953ec125a570545657c651dde8f07adfbd24189372a388ccfe05be4c3e64e57.mp3" }, { "id": "heldout-de-3", "language": "de", "sentences": [ "Jemand hat den kleinen blauen Stuhl umgestellt.", "Er stand hier... neben dem hohen Bücherregal.", "Wir stellen ihn nach dem Essen zurück." ], "text": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier... neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-3", "key": "e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b1c3772aa5d57fab89cadb8c5734f8a85c46b955cc391b0db4a95466eda3ee46", "generation_seconds_batch": 18.070300854975358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.403861999511719, "genuineness": 4.281409740447998, "identity": 0.3251006603240967, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier, neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "ea31e461c643e09664e416887add8910c59dc665f8e913bb1a4268c0ef41146a", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "45239b2859ee3aaf50554d13213489cf596d1c5335071761e0be67ac75a2764a", "generation_seconds_batch": 26.781639675959013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5513381958007812, "target_z": 0.7968749221321973, "ce": 6.125868320465088, "genuineness": 4.111956596374512, "identity": 0.5492178201675415, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ea31e461c643e09664e416887add8910c59dc665f8e913bb1a4268c0ef41146a.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": "ea4d61671f26671636abacedbf2c1c1b3af954a4f46198c20cee935afe566882", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "13e7f2165c4ab4d99cbf427a6c7acf0790dddbce28895341be4f018f625ca400", "generation_seconds_batch": 22.58912706200499, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.964149236679077, "target_z": 1.781250031147121, "ce": 6.3641486167907715, "genuineness": 4.149612903594971, "identity": 0.4617922902107239, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ea4d61671f26671636abacedbf2c1c1b3af954a4f46198c20cee935afe566882.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "ea76c4194a6123252f7631fbc7e9d3c6c563c553724c838c9f646a16d67b154f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0678f181b23f77fbc27a03d860a0e453d3f3eda2dfd1db6b4475a6ac8c2c63e3", "generation_seconds_batch": 25.49970280902926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.132340908050537, "target_z": 1.8984374610660988, "ce": 6.022613525390625, "genuineness": 4.193545341491699, "identity": 0.7058730125427246, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ea76c4194a6123252f7631fbc7e9d3c6c563c553724c838c9f646a16d67b154f.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "eada0b0ea4e2ee09b3e287db45f5e881ccae700410eedbc1414a29fe5f34d8aa", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "296f7b1fdf553e0c5d0f55183d2e817e13abed90e087bebae378e706ec27772a", "generation_seconds_batch": 17.145667833043262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 5.959072113037109, "genuineness": 4.111956596374512, "identity": 0.6018034815788269, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/eada0b0ea4e2ee09b3e287db45f5e881ccae700410eedbc1414a29fe5f34d8aa.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "eba4aacd589e2cb27ca2f51fe192baaf2ae9f116c33ff00ab9771eba7e38435b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "192c0f27dc438f15ef68cb9ebac4e48c992e0b0984f606999e9f0201af2cb5d9", "generation_seconds_batch": 24.12391836300958, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.121128082275391, "target_z": 1.890624932514571, "ce": 6.070269584655762, "genuineness": 3.6789097785949707, "identity": 0.12417742609977722, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/eba4aacd589e2cb27ca2f51fe192baaf2ae9f116c33ff00ab9771eba7e38435b.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "ebd5108adb89539aaf91db55859a63ecf423f7450eb718d5dff940cc699763ee", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "c378c0495babf805755aa302c29677dad465601fcc7aeac978350709242460c4", "generation_seconds_batch": 24.369237206992693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.908085346221924, "target_z": 1.7421875545074619, "ce": 6.324435234069824, "genuineness": 3.986435890197754, "identity": 0.22445708513259888, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ebd5108adb89539aaf91db55859a63ecf423f7450eb718d5dff940cc699763ee.mp3" }, { "id": "heldout-fr-2", "language": "fr", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-2", "key": "ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "357ddab2ec014229d0edbdf1c4dbfd45a1376857580ae4b8285fc87befe750fa", "generation_seconds_batch": 17.17794508393854, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.291382074356079, "target_z": 1.3124999792352525, "ce": 6.141753673553467, "genuineness": 4.507347106933594, "identity": 0.8234724402427673, "hypothesis": "La route tourne à gauche après le pont. Continue, jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "ec050c638bd3bebeb1d6df8172bb40798e8adff62fc842f736539278b514772f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "16f00455d4e6f46779b6c950381c8fbf91095ade7fc901f87086c76e8de5a8f2", "generation_seconds_batch": 23.30425623198971, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7511062622070312, "target_z": 1.6328124870220329, "ce": 6.157639026641846, "genuineness": 4.143336772918701, "identity": 0.39425772428512573, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ec050c638bd3bebeb1d6df8172bb40798e8adff62fc842f736539278b514772f.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": "ec1485656ec5f3e9de706302efc8d3f4548070869109676af105c43610f5a2e5", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "deb34d56f14c151d28be61c836b72f682774666570f826f65db078a47eab38c8", "generation_seconds_batch": 15.64319321908988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.636916160583496, "target_z": 2.2499999169410105, "ce": 6.300607204437256, "genuineness": 4.118232727050781, "identity": 0.5819041728973389, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ec1485656ec5f3e9de706302efc8d3f4548070869109676af105c43610f5a2e5.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "ed6f49a9f0b1619076fa1e7d0ad4e3b0005b957e6c4377b378ab91c9ccae0b9b", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "e7bdcc80a09c9e1e4374c9873a072bd2b46d61325e1a6a6fb88ab02b69f7bcf4", "generation_seconds_batch": 22.79389266099315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.009000301361084, "target_z": 1.8124999792352525, "ce": 6.348263263702393, "genuineness": 4.049196243286133, "identity": 0.5495492219924927, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ed6f49a9f0b1619076fa1e7d0ad4e3b0005b957e6c4377b378ab91c9ccae0b9b.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "ee36910d5632a1b26a0d50ef848ea1ef4a6d6293b04eaa1e573443c300aae0e6", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "dbbf420850a704541d85b1260d43bd3aedb41af195a3291f3e38dba3872d7c44", "generation_seconds_batch": 23.731006611022167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.19046688079834, "target_z": 1.242187388389483, "ce": 6.173524379730225, "genuineness": 4.011539936065674, "identity": 0.662106454372406, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ee36910d5632a1b26a0d50ef848ea1ef4a6d6293b04eaa1e573443c300aae0e6.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "ee4981fb48d1a433b0b82b14e36c0cff42a9b0bcc7e46c837d99090a39d2dbf0", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2453bb21d8d7100ab687258c23d2bdcc7e27ea9f7c26425f5d5e6c94ea4d40c6", "generation_seconds_batch": 32.64235888503026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4147226810455322, "target_z": 1.3984374610660988, "ce": 5.577824115753174, "genuineness": 3.8044304847717285, "identity": 0.460018515586853, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ee4981fb48d1a433b0b82b14e36c0cff42a9b0bcc7e46c837d99090a39d2dbf0.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "ee858d19b3213d3b58db953d7e892e63755d88245713043ee80747d5281af075", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "191813e586cd56df2c02a61a9fc6635b205d7193f1db4859091cbc17afe05c18", "generation_seconds_batch": 24.25161640997976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.941723585128784, "target_z": 1.7656249740440657, "ce": 6.372091293334961, "genuineness": 4.375550270080566, "identity": 0.43346112966537476, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ee858d19b3213d3b58db953d7e892e63755d88245713043ee80747d5281af075.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "eeb758eec389180522f8e54db2a1a8d590a47139e52f7428f0488f6bf963e6e1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "8ad828214fac9f598101c1990d6265b18be82929882b6876d37ec89a914a2ed5", "generation_seconds_batch": 17.62183842493687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.457511901855469, "target_z": 2.1250001245884844, "ce": 5.808161735534668, "genuineness": 3.948779582977295, "identity": 0.41509753465652466, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/eeb758eec389180522f8e54db2a1a8d590a47139e52f7428f0488f6bf963e6e1.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "ef0ca322a176fdd2c543afbbbeeadbdaf83015988d43927e5d0c8b0799e8dcf4", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8c40e8d1e2b49eda3af4807bf309d3cf2feee996b60ca0bf758ab3dd8244a500", "generation_seconds_batch": 17.97186829696875, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.053851127624512, "target_z": 1.8437497612054052, "ce": 5.990842819213867, "genuineness": 3.835810661315918, "identity": 0.5628300905227661, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ef0ca322a176fdd2c543afbbbeeadbdaf83015988d43927e5d0c8b0799e8dcf4.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": "ef3757a828f2c800ead87635ddba58ca15fd6719dbef3df1e074829924bd625f", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab560e38d65141533a8bad0511a92e1b4dfcd628be82b2e0f728629421f72e57", "generation_seconds_batch": 18.17393783503212, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.098702430725098, "target_z": 1.8749998754115158, "ce": 6.125868320465088, "genuineness": 4.055472373962402, "identity": 0.739189863204956, "hypothesis": "Ik schreef het nummer op de envelop, lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ef3757a828f2c800ead87635ddba58ca15fd6719dbef3df1e074829924bd625f.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "efb7b289ce2fe9b916c9f3845f9613c6657fac8097e74979922987c48c9e6861", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "98e43fa1a53c46bb0f9c9d4d7a51e655a0ada7835700bd9551565dc6c6c9b371", "generation_seconds_batch": 23.160408871015534, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8989346027374268, "target_z": 1.0390624766396592, "ce": 6.0305562019348145, "genuineness": 3.7918784618377686, "identity": 0.37273257970809937, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/efb7b289ce2fe9b916c9f3845f9613c6657fac8097e74979922987c48c9e6861.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "f0082ee7338edeb4054d1745f8ddc19f611b05b4927d4b6b091cd3dcdea8df87", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "842a2fb96dbb8f69ee42022302bde7ccc23f6962e5ee220d55b2e28344b9cad1", "generation_seconds_batch": 24.634088345919736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5604889392852783, "target_z": 1.5, "ce": 6.221180438995361, "genuineness": 3.9801597595214844, "identity": 0.6811286211013794, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f0082ee7338edeb4054d1745f8ddc19f611b05b4927d4b6b091cd3dcdea8df87.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "f073962d5935bb2a1e0b996a69dd34de00cf53797588921f05b40f78d74f9b6c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "594e598f2ae139caf45975192b628ac4fe0051aa71e6a95f66c6c71b2ffcd32f", "generation_seconds_batch": 23.152368337963708, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.986574649810791, "target_z": 1.7968749221321973, "ce": 6.1894097328186035, "genuineness": 4.068024158477783, "identity": 0.492040753364563, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f073962d5935bb2a1e0b996a69dd34de00cf53797588921f05b40f78d74f9b6c.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "f0b6625c736bcbb856bf09215648da79c3504b65a0c3f13df4ec797471cc6dae", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "02f4f5f392dff18b0c3325bb2060ef5a7760dc347ddb92a33a73779a73f500b2", "generation_seconds_batch": 26.718650197959505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5829145908355713, "target_z": 1.5156250571030552, "ce": 6.284721851348877, "genuineness": 4.319066047668457, "identity": 0.46366506814956665, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f0b6625c736bcbb856bf09215648da79c3504b65a0c3f13df4ec797471cc6dae.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "f0cdaa91be0f0ee9225ced419d4463f8c899c12c72b0f916f310d129131860f7", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "39de24624ec5d1fdb71d29922a08aef37b329d3bf1cebddab40f187edd518fb2", "generation_seconds_batch": 26.84770915308036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165979385375977, "target_z": 1.9218750467206815, "ce": 5.967014789581299, "genuineness": 3.9173994064331055, "identity": 0.4321976602077484, "hypothesis": "La puerta sigue abierta. Te he dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.13333333333333333, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f0cdaa91be0f0ee9225ced419d4463f8c899c12c72b0f916f310d129131860f7.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "f124162d8fcaebdda84e2d156f2c29712904b366f44bb0f3e51e2b8168d49095", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "799e4b19909439515862d763fdc50cc0c345847cea9a7f2bfe8abec85cc2dc4c", "generation_seconds_batch": 23.861214244971052, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.031425952911377, "target_z": 1.828125036338308, "ce": 6.125868320465088, "genuineness": 4.111956596374512, "identity": 0.6985160112380981, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f124162d8fcaebdda84e2d156f2c29712904b366f44bb0f3e51e2b8168d49095.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "82f1a174e94cd598220b187b28521f50b142d303d8bb0d31d07c2cca9401df2a", "generation_seconds_batch": 16.61897993204184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.627765655517578, "target_z": 1.5468750051911868, "ce": 6.1894097328186035, "genuineness": 4.319066047668457, "identity": 0.858466625213623, "hypothesis": "Ton message est arrivé après mon départ, je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "f256e864f20a51a7f8d56bbc67cd8d1681c1feaa07d680a2f5f148322292a3f2", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d81d3f22f71a6bc2c21f276fbdf423d520a509e2bdfb9098635e89f30a1eebdf", "generation_seconds_batch": 25.46429295593407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.278107166290283, "target_z": 2.0, "ce": 6.3641486167907715, "genuineness": 4.068024158477783, "identity": 0.3844951391220093, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f256e864f20a51a7f8d56bbc67cd8d1681c1feaa07d680a2f5f148322292a3f2.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "f2c78a32e2ef6f6130fe062e0746ad4fd0df615f5d08b05bddb5e9d6eace7cfc", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "5fb8f2bce55ba88c257e8c3e4d89030ab39c01fcee5ebad515505fff1f57c056", "generation_seconds_batch": 23.70501429901924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.457511901855469, "target_z": 2.1250001245884844, "ce": 6.3641486167907715, "genuineness": 3.8922953605651855, "identity": 0.4372596740722656, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f2c78a32e2ef6f6130fe062e0746ad4fd0df615f5d08b05bddb5e9d6eace7cfc.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "f2da2019b61fbcf6416cd22b1b4119a258746f020ee571f8a26eb654ed20f76a", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "189516f2ba10e093919597f8d355e14232a0abc96325c8221d8ba2274e14283c", "generation_seconds_batch": 23.517710080952384, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.031425952911377, "target_z": 1.828125036338308, "ce": 6.1020402908325195, "genuineness": 3.537698984146118, "identity": 0.1729930341243744, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f2da2019b61fbcf6416cd22b1b4119a258746f020ee571f8a26eb654ed20f76a.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "f34214c1fe898140cc43af567801c8a9af200f09b7b0d9cf07d2df26e7540059", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "193eaae57889e58f54d924abcef2763d423e311e04fe290edbea5d3a8c1e811f", "generation_seconds_batch": 24.346117430948652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.672616720199585, "target_z": 1.5781249532793185, "ce": 6.23706579208374, "genuineness": 4.20609712600708, "identity": 0.6679048538208008, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f34214c1fe898140cc43af567801c8a9af200f09b7b0d9cf07d2df26e7540059.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "f3797e6f05827aea5d05aef3b9349e44e2743b922e8db9c02cd1a1167d941d6c", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "89d6a5961085b0e87cc20aed75c7f5ad20674e683ccb0b61581093b12599cbb9", "generation_seconds_batch": 18.041863624937832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5829145908355713, "target_z": 1.5156250571030552, "ce": 5.442798614501953, "genuineness": 3.8797430992126465, "identity": 0.47976183891296387, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f3797e6f05827aea5d05aef3b9349e44e2743b922e8db9c02cd1a1167d941d6c.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "f5365823c38b160759c790f36f2cddb6284039ac2c05039df8aa81eb4ba0df3e", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "729d1cb5eed486cc54329bed6eb3e84f0fdb965e46768d6a3097cdf458c3c2b8", "generation_seconds_batch": 21.40313338499982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 5.8875885009765625, "genuineness": 3.9299514293670654, "identity": 0.6019338965415955, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f5365823c38b160759c790f36f2cddb6284039ac2c05039df8aa81eb4ba0df3e.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "f5409c44a5c532b1114e8b29c3dd3d8528447bf4c5effb68d403d85e4f486bdc", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f414bfeab7722a9ed9df6f1ca3d63ca2062b41c62149b70de83de4cd5bf58e86", "generation_seconds_batch": 26.765301414066926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.773531913757324, "target_z": 1.6484375441250882, "ce": 6.165581703186035, "genuineness": 4.231201171875, "identity": 0.7761092185974121, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f5409c44a5c532b1114e8b29c3dd3d8528447bf4c5effb68d403d85e4f486bdc.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "f5d35d0d08b25a3fecca6ddfa66f49486a8a925ce215cf4c6d19ed2a58252908", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "eb187794822e09a08b196de525e1d0f4685d455786128381620f15021c9d72fe", "generation_seconds_batch": 23.970778897986747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.549276113510132, "target_z": 1.4921874714484724, "ce": 6.054384231567383, "genuineness": 3.998987913131714, "identity": 0.5158874988555908, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f5d35d0d08b25a3fecca6ddfa66f49486a8a925ce215cf4c6d19ed2a58252908.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "f5de5daecef49c65607feac0c6813918dc694416b41c19834d6e25a321206e83", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "5437c1068658ea9e424b88431dd9b78eaf112c5ba96f95aefe76efeb40718ea1", "generation_seconds_batch": 17.295295965042897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.502363204956055, "target_z": 2.1562502387945948, "ce": 6.173524379730225, "genuineness": 4.20609712600708, "identity": 0.6509121656417847, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f5de5daecef49c65607feac0c6813918dc694416b41c19834d6e25a321206e83.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": "f6079286cf7f3b607adcae15094fd80f12cea18eb90393938e025bbbe79a33da", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad0fb6b97f63eac2fd698b0e6a685fa558bd2fdb27467e679aa7e774ee379041", "generation_seconds_batch": 23.234204292064533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.076276779174805, "target_z": 1.8593748183084604, "ce": 6.1894097328186035, "genuineness": 3.9801597595214844, "identity": 0.5641891360282898, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f6079286cf7f3b607adcae15094fd80f12cea18eb90393938e025bbbe79a33da.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "f631b8ce6f5bbff5f67caa69f4013f8590f04829d64b9ce8e3d69390a267649d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "53c6bd79102bc91a6b7ba79831dfd5af600bad619e0940e8488c41ec13a77987", "generation_seconds_batch": 25.49970280902926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.885659694671631, "target_z": 1.7265624974044065, "ce": 6.205295085906982, "genuineness": 4.13078498840332, "identity": 0.6917134523391724, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f631b8ce6f5bbff5f67caa69f4013f8590f04829d64b9ce8e3d69390a267649d.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": "f6e321f38e70a81991a4bf42303d7a79ad31e6690011e5a903876c047d2216e2", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "217f14f307f9cb1660bf9bb5aea988516149e847aedcb3288eafdc7b9eaa702e", "generation_seconds_batch": 22.407566937967204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.076276779174805, "target_z": 1.8593748183084604, "ce": 6.09409761428833, "genuineness": 4.143336772918701, "identity": 0.6909623742103577, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f6e321f38e70a81991a4bf42303d7a79ad31e6690011e5a903876c047d2216e2.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "f70f9d98e5252a943f5b403a4f72983f855f9d35d5de5af7eeb5f64e97dfb9b9", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "e5459f5c228da53c8142a3c88c03ae0b8e6979408cd8113fbd73fc5dd93d8b3f", "generation_seconds_batch": 23.936171105946414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9192981719970703, "target_z": 1.7500000830589895, "ce": 6.0305562019348145, "genuineness": 3.9927117824554443, "identity": 0.599216639995575, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f70f9d98e5252a943f5b403a4f72983f855f9d35d5de5af7eeb5f64e97dfb9b9.mp3" }, { "id": "vn_S_NARR_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-1", "key": "f7a108e6f5b6eb895f5ed5b15f9f8a55a02959b0e2cf8717cd7d2a48770d0eae", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] I found the little box.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] It was beside the old window.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "a5f510d23428fc3bb1165203202f0f4724aad36c48cf08715fd1f45b918fb1b8", "generation_seconds_batch": 22.345885685062967, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0202131271362305, "target_z": 1.8203125077867803, "ce": 6.268836498260498, "genuineness": 4.2374773025512695, "identity": 0.47770747542381287, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f7a108e6f5b6eb895f5ed5b15f9f8a55a02959b0e2cf8717cd7d2a48770d0eae.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "480eb50a310d698e8f5adf13fd505311fd6f8a3b52dba6df2c92aa8d6bde315f", "generation_seconds_batch": 18.033019082038663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.098702430725098, "target_z": 1.8749998754115158, "ce": 6.086154937744141, "genuineness": 4.394378662109375, "identity": 0.7424219846725464, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "f82446a26d7563fe63620a261708ef41d110d1244ba5b4b9b3dde6c8da5e52b3", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "01905f5b1c3edf644ec03145640cdb084c2db40b8a4aeaf58f5ddaef10b92d4e", "generation_seconds_batch": 17.53504059801344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3698716163635254, "target_z": 1.3671875129779671, "ce": 6.086154937744141, "genuineness": 3.7918784618377686, "identity": 0.19164952635765076, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f82446a26d7563fe63620a261708ef41d110d1244ba5b4b9b3dde6c8da5e52b3.mp3" }, { "id": "heldout-de-2", "language": "de", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-2", "key": "f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fda0694828d8ab6e3949d92dbc2c2688722c003cd7205e84e363e8b7566a99ed", "generation_seconds_batch": 18.278840276994742, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.278107166290283, "target_z": 2.0, "ce": 6.221180438995361, "genuineness": 4.3881025314331055, "identity": 0.6221126317977905, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "f87e4c4ab48703e7239cf0251c9e6c755ef335a4f838264a1b0f657e2bb1a2f2", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "1ba811ea4341c77775265dbe189d12388bf5a2ae4af8b37cebd21d5eca71df3f", "generation_seconds_batch": 23.66782324796077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165979385375977, "target_z": 1.9218750467206815, "ce": 6.149696350097656, "genuineness": 3.427868366241455, "identity": 0.2550174593925476, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f87e4c4ab48703e7239cf0251c9e6c755ef335a4f838264a1b0f657e2bb1a2f2.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "f8fefa58976055d1cbc623062376bea028b4150f161b8c9535fdc3a0d0cf9fb2", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f841200400a628b28a98cdeb78994d869cb4751d1d7cfb833744c0a71422b9a", "generation_seconds_batch": 17.295295965042897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.062326908111572, "genuineness": 4.218649387359619, "identity": 0.5926534533500671, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f8fefa58976055d1cbc623062376bea028b4150f161b8c9535fdc3a0d0cf9fb2.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "f90dd3593cdb8e04fe8cd8b532d5d61fedd98c1bb6bad76eed1068b22c47567d", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "c1580ed2fbb0ffaf2cc953b4c51e94859165fcfa8cdf085310a5d17dd4fc016f", "generation_seconds_batch": 16.677778678014874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.526850700378418, "target_z": 1.4765625804633962, "ce": 6.0146708488464355, "genuineness": 4.118232727050781, "identity": 0.43008917570114136, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f90dd3593cdb8e04fe8cd8b532d5d61fedd98c1bb6bad76eed1068b22c47567d.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "f9548f4d81e20c9137eaad87570f99c5962403e28545ecd7e272b75150b43566", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "c62e2dc7c599396cc00db7aa6ede42178e43f436d4597061d0cadc7bec2e3a8d", "generation_seconds_batch": 24.69295376900118, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25568151473999, "target_z": 1.9843749428969448, "ce": 6.141753673553467, "genuineness": 4.300237655639648, "identity": 0.6387283802032471, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f9548f4d81e20c9137eaad87570f99c5962403e28545ecd7e272b75150b43566.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "f9db326d82a31839ddcd4fa316ea780758bd285d80f2ca1cecdf548b987f50a8", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c0c3f37ac2938397153b44dbb4c3683eb60c63dcad32d6dcbec7f15c62fddf9d", "generation_seconds_batch": 16.348945526056923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.322957992553711, "target_z": 2.0312497819701525, "ce": 6.125868320465088, "genuineness": 3.998987913131714, "identity": 0.6932278275489807, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/f9db326d82a31839ddcd4fa316ea780758bd285d80f2ca1cecdf548b987f50a8.mp3" }, { "id": "vn_S_NARR_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-en-0", "key": "faea17d8f36b09c4f86e5309b258a440a3b563f68d75c873ccd98b04ae2818a3", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] The door is still open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] I put your coat on the chair.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "973b2f95f21e0eb82f5bd4d062531ccf98289d3b27a5e0a7a0492a9a73d4b30d", "generation_seconds_batch": 21.853655782993883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.336233139038086, "target_z": 1.3437499273233842, "ce": 6.109982967376709, "genuineness": 3.9550557136535645, "identity": 0.2198495864868164, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/faea17d8f36b09c4f86e5309b258a440a3b563f68d75c873ccd98b04ae2818a3.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "fb28cc828f4e72054b5029e2bf3fb2f3b04a42ceb4f786451740c7aa2661b434", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "0d87ef4226812da05fd4bca2412b44fb78f577ea218b6fb31c576d9e172defe2", "generation_seconds_batch": 23.982853319961578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.941723585128784, "target_z": 1.7656249740440657, "ce": 6.09409761428833, "genuineness": 4.017816066741943, "identity": 0.6976988315582275, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/fb28cc828f4e72054b5029e2bf3fb2f3b04a42ceb4f786451740c7aa2661b434.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "fb9efaf389b5821f38604c1dac9062ef37e92221246f0849c503b220e8490031", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "e2a7904f45b88aeeb5e1e0708a40605b2cbe194ed1822e6e1630e2b67e31f12f", "generation_seconds_batch": 24.190641853027046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.547214031219482, "target_z": 2.1875000207647473, "ce": 6.324435234069824, "genuineness": 3.8295347690582275, "identity": 0.5267006158828735, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/fb9efaf389b5821f38604c1dac9062ef37e92221246f0849c503b220e8490031.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "fc012d234b6e5d7a5e6244cf5a330d8b3a44cac8e645aac835b1ba33642472e1", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 127, "seconds_generated": 10.16, "mp3_sha256": "931f3eeff03173add49d7ace5922fa27ec07e8eb3e2b6bf19d3775a284e692c4", "generation_seconds_batch": 17.782641093013808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.614490509033203, "target_z": 2.2343748598379554, "ce": 6.2767791748046875, "genuineness": 4.268857479095459, "identity": 0.6101306676864624, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/fc012d234b6e5d7a5e6244cf5a330d8b3a44cac8e645aac835b1ba33642472e1.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "fc5266c5b9eb0fafb37f3cec7021e53eb8e1b7f5a981339a0caeaf96db71b215", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8a26cc56fa4728d99957d957f4bd02bc81b881b6b38c4dd01a6149decbd1b41d", "generation_seconds_batch": 30.98853924300056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7847447395324707, "target_z": 1.6562500726766158, "ce": 6.0305562019348145, "genuineness": 4.080576419830322, "identity": 0.5655666589736938, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/fc5266c5b9eb0fafb37f3cec7021e53eb8e1b7f5a981339a0caeaf96db71b215.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "fc7a27aad5bf7208b63418e9a510509cd99c1bd446f1c53e53dcf98afe052161", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "b54ae8e7a301a685143cd9b7dc3cbec0c3dcb73d3e1dd13d933ed18bcf36fff4", "generation_seconds_batch": 27.131078917067498, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.773531913757324, "target_z": 1.6484375441250882, "ce": 5.982900142669678, "genuineness": 3.973883628845215, "identity": 0.4941878318786621, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/fc7a27aad5bf7208b63418e9a510509cd99c1bd446f1c53e53dcf98afe052161.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "fc7be57e0dcaffb8bb820a5fe6a88a74b8c27dcf7b7f5c9e21e6181165321969", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "638349020136de5f00b3a103f82a1e2669ec746ceaa6b1f23469a94d51a5c290", "generation_seconds_batch": 23.99903186794836, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8744468688964844, "target_z": 1.718749968852879, "ce": 6.24500846862793, "genuineness": 4.093128204345703, "identity": 0.6558831930160522, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/fc7be57e0dcaffb8bb820a5fe6a88a74b8c27dcf7b7f5c9e21e6181165321969.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": "fc80ae677eabec4b724d72d488c0ca1ef1f45ecd9aa52e7b0b0ca810e03892c2", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6fbcc239656c730a5bca6393321855038cee84dbde38b4c58481868998a805fb", "generation_seconds_batch": 17.951629588962533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.222043037414551, "target_z": 1.9609373572423618, "ce": 5.974957466125488, "genuineness": 4.118232727050781, "identity": 0.6452020406723022, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/fc80ae677eabec4b724d72d488c0ca1ef1f45ecd9aa52e7b0b0ca810e03892c2.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "fcf03b9ce0d94acffc449ccb40ff860f8cd74b2ddc39c20cc4f8a44a54f6d7e4", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "fb7f513faa1d95dac0a60ad2fc0211cf3e34c1c2299dbd571c649b7d743c4cd8", "generation_seconds_batch": 23.605451787938364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.076276779174805, "target_z": 1.8593748183084604, "ce": 6.213237762451172, "genuineness": 4.093128204345703, "identity": 0.711685299873352, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/fcf03b9ce0d94acffc449ccb40ff860f8cd74b2ddc39c20cc4f8a44a54f6d7e4.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": "fcf8ecd6bb9bce31a02d07bc26268c6cc9fbb862205d984b8523158771f96832", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "627d7389cdedd3f10e6bf2c1e1dc18d3ceb7fcf075ccf61d43ee0cd51781cafe", "generation_seconds_batch": 21.532504761009477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7062551975250244, "target_z": 1.6015625389339012, "ce": 6.157639026641846, "genuineness": 4.1998209953308105, "identity": 0.666073203086853, "hypothesis": "Alguien ha movido la pequeña silla azul. ¿Estaba aquí? Junto a la estantería alta la colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/fcf8ecd6bb9bce31a02d07bc26268c6cc9fbb862205d984b8523158771f96832.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "fd1865ea313bb7878a04fd7197540a98fad06e57a70287e725a214ef48a4b8f5", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "e7851d3ad8cbc64c906ca05589bbf7d5ca23dee9b2ecc04d29d78b2fee502747", "generation_seconds_batch": 23.02787744498346, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7847447395324707, "target_z": 1.6562500726766158, "ce": 6.09409761428833, "genuineness": 4.080576419830322, "identity": 0.7307482957839966, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/fd1865ea313bb7878a04fd7197540a98fad06e57a70287e725a214ef48a4b8f5.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": "fd2fb0070e8d71642176f3c1d530904764fed37a0117bcff3536c2416758e736", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2b5ea37b954822afb2833e82670cc8060ef194f4a7c8c0aaaa37ae72af1a496b", "generation_seconds_batch": 16.375146290054545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.549276113510132, "target_z": 1.4921874714484724, "ce": 6.41180419921875, "genuineness": 4.319066047668457, "identity": 0.513858437538147, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/fd2fb0070e8d71642176f3c1d530904764fed37a0117bcff3536c2416758e736.mp3" }, { "id": "vn_S_NARR_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-1", "key": "fddfbf4ac60ce5a688140e613af0daadc37d6d2d9ba9c4ef3d222b17062d1d12", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "bdb855e8aa71c9502f1b7462e16b58949248ca60e98843d53b0c36ce74165bd9", "generation_seconds_batch": 29.972136833006516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7623190879821777, "target_z": 1.6406250155735604, "ce": 5.764476776123047, "genuineness": 4.21237325668335, "identity": 0.47473907470703125, "hypothesis": "Mouvi. J'ai 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": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/fddfbf4ac60ce5a688140e613af0daadc37d6d2d9ba9c4ef3d222b17062d1d12.mp3" }, { "id": "vn_S_NARR_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-fr-0", "key": "fe533346fc1d4aa1410cd1cdb02ad4670c525346137e755a12d30d01d4a25be0", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "45e65ebc157615b1421ce640ed8b203f45f58dcc9f3e8d82867bbae6b3d353b1", "generation_seconds_batch": 24.055840765009634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.795957326889038, "target_z": 1.6640624351101645, "ce": 6.109982967376709, "genuineness": 3.9676074981689453, "identity": 0.6668286323547363, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/fe533346fc1d4aa1410cd1cdb02ad4670c525346137e755a12d30d01d4a25be0.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "fe6a2beac6044aa966cb8c77a2e1536343105e10ecad7c33f37fcac29ff338ed", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "f8294d298d49cc26c3081fbe4684346a27f9340281f962acdcbfaff94aa65a46", "generation_seconds_batch": 23.539477342972532, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.986574649810791, "target_z": 1.7968749221321973, "ce": 6.125868320465088, "genuineness": 4.124508857727051, "identity": 0.22347041964530945, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/fe6a2beac6044aa966cb8c77a2e1536343105e10ecad7c33f37fcac29ff338ed.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "fe7c63e4039ff38c0670a0373329b5deee8de71496970298b2e09b3658ca7c5f", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "59870013b15089c47943df1114fb496972f84d76182be5f48260582080d37c94", "generation_seconds_batch": 24.479775142041035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.704193115234375, "target_z": 2.2968750882501765, "ce": 6.23706579208374, "genuineness": 4.18099308013916, "identity": 0.6563490629196167, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/fe7c63e4039ff38c0670a0373329b5deee8de71496970298b2e09b3658ca7c5f.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "febf6df7a9e266a849ef0caea0baf1db4e6131a39e5f26bfbf791cf9fa69afba", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "94ce4ffe4f230884052572945a69b98ce5991a75e7e01d28c5166f26ed07a6eb", "generation_seconds_batch": 15.91696778696496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.986574649810791, "target_z": 1.7968749221321973, "ce": 6.054384231567383, "genuineness": 3.7542223930358887, "identity": -0.03674812614917755, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/febf6df7a9e266a849ef0caea0baf1db4e6131a39e5f26bfbf791cf9fa69afba.mp3" }, { "id": "vn_S_NARR_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-0", "key": "fef741be7595da0db9ec87ee18d02bf58ebb04919ab958f8c147c0e145b70384", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "0b50e42764864fd9898483a588b892c72bbf84f67025d852efe973606c190fd7", "generation_seconds_batch": 18.01036707102321, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.403510093688965, "target_z": 1.3906250986325501, "ce": 5.903473377227783, "genuineness": 3.9927117824554443, "identity": 0.327345073223114, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/fef741be7595da0db9ec87ee18d02bf58ebb04919ab958f8c147c0e145b70384.mp3" }, { "id": "vn_S_NARR_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-de-1", "key": "ff054e129506b8fd32af07c307cf057fbd5edcb4fbee78ddd57bc9e28bb20ad8", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ec3a3e897543a48d62110801d4c6085d62e15f0796c3946694159bc87a1fad9c", "generation_seconds_batch": 16.915298811043613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2108306884765625, "target_z": 1.9531251609267921, "ce": 6.205295085906982, "genuineness": 3.6789097785949707, "identity": 0.11863243579864502, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ff054e129506b8fd32af07c307cf057fbd5edcb4fbee78ddd57bc9e28bb20ad8.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": "ff27dc6dc3cad3101c2c91a4377c2339c6c7f45cadd774f97380b890dafcc7e7", "bucket": "vn_S_NARR_reg", "phase": "heldout", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e97e44e10d07d53b1b2531623fc3b56397bd9f2a46b3ab812fece6d002819016", "generation_seconds_batch": 20.8017126879422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.941723585128784, "target_z": 1.7656249740440657, "ce": 6.109982967376709, "genuineness": 4.149612903594971, "identity": 0.4899023175239563, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ff27dc6dc3cad3101c2c91a4377c2339c6c7f45cadd774f97380b890dafcc7e7.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "ff32c44788e5b0a833994491c4e7e378b3f62a49475200d969ba796079e58caa", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "b71a16858170c21cfcabd28a139e8c354a92bb1624a98c0b2541eb2f164dec53", "generation_seconds_batch": 22.904354044003412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7623190879821777, "target_z": 1.6406250155735604, "ce": 6.292664527893066, "genuineness": 4.168440818786621, "identity": 0.6007463335990906, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ff32c44788e5b0a833994491c4e7e378b3f62a49475200d969ba796079e58caa.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "ff52a88ee8c884dfe2f062514cacdf944fe80e276f0f9aa6380358a31d4e9af7", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "fc82c6264295e4025b7b1288ae91734d0902f49a7a937c09cad26763a0d0cc12", "generation_seconds_batch": 22.193334951996803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8744468688964844, "target_z": 1.718749968852879, "ce": 6.252951145172119, "genuineness": 4.074300289154053, "identity": 0.6666361093521118, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ff52a88ee8c884dfe2f062514cacdf944fe80e276f0f9aa6380358a31d4e9af7.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "ff55701d2746ec580252be6439c66860cd0b06efb73156ee950cb9e1f2965c67", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dbb07ad0165b7a8903ce42af63ce006041fe2146fdc410b3f7dae9986066fa11", "generation_seconds_batch": 23.471925292047672, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.95499849319458, "target_z": 1.0781249532793185, "ce": 5.998785495758057, "genuineness": 3.8797430992126465, "identity": 0.6761811971664429, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ff55701d2746ec580252be6439c66860cd0b06efb73156ee950cb9e1f2965c67.mp3" }, { "id": "vn_S_NARR_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-0", "key": "ff787b72b07ef2ac2e62cc845bc9f382f39e8cd4ce0ec7402b2fb534ee600f85", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "2f084851ea20eb906be370ead320c9d89c420d2e1d85bb49e2b17b90e8c354ba", "generation_seconds_batch": 24.317695157951675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5247883796691895, "target_z": 2.171874963661692, "ce": 6.300607204437256, "genuineness": 3.948779582977295, "identity": 0.5134921073913574, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ff787b72b07ef2ac2e62cc845bc9f382f39e8cd4ce0ec7402b2fb534ee600f85.mp3" }, { "id": "vn_S_NARR_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-1", "key": "ffc04c61d4c47236bb8f9189dc378b45a7bb828991bf16eb47d36ce3f3dad458", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "18e978606726bb82e9b649aebf472f27fecd5376720b55cfaf4144cf429aae45", "generation_seconds_batch": 22.217977733002044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.065064430236816, "target_z": 1.851562621992891, "ce": 6.205295085906982, "genuineness": 4.099404335021973, "identity": 0.5852831602096558, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ffc04c61d4c47236bb8f9189dc378b45a7bb828991bf16eb47d36ce3f3dad458.mp3" }, { "id": "vn_S_NARR_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-nl-0", "key": "ffcd3f7df403e5b581dbd88c781fd2c922ebfa686381b2bf067ec2d16514d439", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] De deur staat nog open.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "ecf0a107453f6fd5486a7373e7fbfdf85e83d18a65b9e4c247655222f97b2438", "generation_seconds_batch": 18.055538668995723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3698716163635254, "target_z": 1.3671875129779671, "ce": 6.006728172302246, "genuineness": 4.093128204345703, "identity": 0.564116358757019, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ffcd3f7df403e5b581dbd88c781fd2c922ebfa686381b2bf067ec2d16514d439.mp3" }, { "id": "vn_S_NARR_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_NARR_reg-es-1", "key": "ffe0abd5afb0d0877d094d3bc4c9ea4ab35d75e81dc4d5d5668be2b74bfb21d0", "bucket": "vn_S_NARR_reg", "phase": "grid", "candidate": "manual-9ea232963c9d", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, ensuring every consonant is cleanly formed while maintaining a relaxed jaw posture.\nSCRIPT:\n(Deliver the opening segment with crisp, clear articulation, emphasizing the initial consonant sounds.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-section demands meticulous enunciation of all mid-range vowels and connecting sounds.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with sharp, defined final consonants, giving the narration a clean, authoritative stop.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "867accbc99e6355bbee51923c914d05b50caa831d952e151238572b8d02d7bde", "generation_seconds_batch": 24.337888716952875, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.502363204956055, "target_z": 2.1562502387945948, "ce": 6.070269584655762, "genuineness": 4.218649387359619, "identity": 0.609730064868927, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_NARR_reg/ffe0abd5afb0d0877d094d3bc4c9ea4ab35d75e81dc4d5d5668be2b74bfb21d0.mp3" }, { "key": "original-vn_S_NARR_reg-de-0-seed1701-scale0.0", "bucket": "vn_S_NARR_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "755682df5c7f5570428075c03fb9dcf0c8557c69cdf22885336091b2cee841e6", "task_id": "vn_S_NARR_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.672616720199585, "target_z": 1.5781249532793185, "ce": 6.332377910614014, "genuineness": 4.344170093536377, "identity": 0.48723942041397095, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/de/525afc0b6a5199dbba2da931.mp3" }, { "key": "original-vn_S_NARR_reg-de-0-seed1701-scale0.5", "bucket": "vn_S_NARR_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "4e6edba46cb1ef65ce01000e963f685be0da583333eb7bf64345550801c2825d", "task_id": "vn_S_NARR_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.695042371749878, "target_z": 1.5937500103823736, "ce": 6.173524379730225, "genuineness": 4.457139015197754, "identity": 0.5326755046844482, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/de/85c3218d22f1e25b7e4fe895.mp3" }, { "key": "original-vn_S_NARR_reg-de-0-seed1701-scale0.75", "bucket": "vn_S_NARR_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "75ffb664b1fa5b6e707be7bf809c702b9686d359de921bf5544f6b1dfbe593a6", "task_id": "vn_S_NARR_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1568286418914795, "target_z": 1.218749968852879, "ce": 6.252951145172119, "genuineness": 4.394378662109375, "identity": 0.552635133266449, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/de/09ab41dd0b5f236989de15d7.mp3" }, { "key": "original-vn_S_NARR_reg-de-0-seed1701-scale1.0", "bucket": "vn_S_NARR_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "feef6eeba00390403d463b93e0526842e84071d0bed6f58218627581eacc011a", "task_id": "vn_S_NARR_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.885659694671631, "target_z": 1.7265624974044065, "ce": 6.340320587158203, "genuineness": 4.3504462242126465, "identity": 0.4535507559776306, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/de/fdd14d28820ebff01a4eb03d.mp3" }, { "key": "original-vn_S_NARR_reg-de-0-seed1702-scale0.0", "bucket": "vn_S_NARR_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "25708113d9e60ca657b3e3ed48c492bb5110e1bb7dd7fd00917e9850059d7958", "task_id": "vn_S_NARR_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.358658790588379, "target_z": 1.3593749844264396, "ce": 6.205295085906982, "genuineness": 4.457139015197754, "identity": 0.641600489616394, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/de/0f2379c715f9ad82f5c91a75.mp3" }, { "key": "original-vn_S_NARR_reg-de-0-seed1702-scale0.5", "bucket": "vn_S_NARR_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "3aa3bcd7bcad9820445454b36954d370fb80df7c7b8b6f4eb43e2194adde9780", "task_id": "vn_S_NARR_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5380632877349854, "target_z": 1.4843749428969448, "ce": 6.133810997009277, "genuineness": 4.243753433227539, "identity": 0.5763013362884521, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/de/141d7f5cb4db8844e5a35109.mp3" }, { "key": "original-vn_S_NARR_reg-de-0-seed1702-scale0.75", "bucket": "vn_S_NARR_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "3fb9e894dd6652d2eb5b0029220cf52d6dd9c138cc022ded15730e3537791716", "task_id": "vn_S_NARR_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8744468688964844, "target_z": 1.718749968852879, "ce": 6.229123115539551, "genuineness": 4.3127899169921875, "identity": 0.4532548189163208, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/de/46c9c4ce6d51075b2a922e85.mp3" }, { "key": "original-vn_S_NARR_reg-de-0-seed1702-scale1.0", "bucket": "vn_S_NARR_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "b777aac2fb460939678bedf5fb679eaffc21ad915f6a6a44dd28db02b5235148", "task_id": "vn_S_NARR_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1568286418914795, "target_z": 1.218749968852879, "ce": 6.1894097328186035, "genuineness": 4.469690799713135, "identity": 0.7089284658432007, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/de/b1c6cac81d49f6dec7c48862.mp3" }, { "key": "original-vn_S_NARR_reg-de-1-seed1701-scale0.0", "bucket": "vn_S_NARR_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "3bbc80443df2c9302f247fbf47bd0f675d8755078d0093188265b7cf042e9be8", "task_id": "vn_S_NARR_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.479937553405762, "target_z": 2.1406251816915396, "ce": 6.38003396987915, "genuineness": 4.262581825256348, "identity": 0.5049885511398315, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/de/0ec3406f411979dbbdbfda6a.mp3" }, { "key": "original-vn_S_NARR_reg-de-1-seed1701-scale0.5", "bucket": "vn_S_NARR_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "a99e3457cc14b829ecae3eab7ec3ead0f2d3b69bd521efe36d6fbf9e02ef7992", "task_id": "vn_S_NARR_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.39023494720459, "target_z": 2.0781249532793185, "ce": 6.4515180587768555, "genuineness": 4.494794845581055, "identity": 0.5527129173278809, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster, wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/de/8d96801f1f12b4e1946081ef.mp3" }, { "key": "original-vn_S_NARR_reg-de-1-seed1701-scale0.75", "bucket": "vn_S_NARR_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "ec60d1cb399c20a58c5db441f83312ed6150b00b8828bf7c5d20200412ad7f82", "task_id": "vn_S_NARR_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.457511901855469, "target_z": 2.1250001245884844, "ce": 6.435632705688477, "genuineness": 4.168440818786621, "identity": 0.5534130930900574, "hypothesis": "Ich habe die kleine Schacht gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/de/d87da91000f37e9f023fe7bf.mp3" }, { "key": "original-vn_S_NARR_reg-de-1-seed1701-scale1.0", "bucket": "vn_S_NARR_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "231fd3c6b037b9fd89cfc8b2118c78d211119c823c3963c84a33fa677b7edeb3", "task_id": "vn_S_NARR_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2108306884765625, "target_z": 1.9531251609267921, "ce": 6.316492557525635, "genuineness": 4.036643981933594, "identity": 0.6450326442718506, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/de/fd48439b8b78fa3bdcdb7b00.mp3" }, { "key": "original-vn_S_NARR_reg-de-1-seed1702-scale0.0", "bucket": "vn_S_NARR_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "1f40aacf5fe358bde85b31e6fc6c7b3cb48c31b2e323010cae21f5c28bb9b544", "task_id": "vn_S_NARR_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.367809295654297, "target_z": 2.0624998961762633, "ce": 6.316492557525635, "genuineness": 4.13078498840332, "identity": 0.42938196659088135, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/de/a29cdcc5092cc3b727da4963.mp3" }, { "key": "original-vn_S_NARR_reg-de-1-seed1702-scale0.5", "bucket": "vn_S_NARR_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "d7573d6d3705cef447f5efc0f6c5d86dafe2178ba7a8413da8220f7f42bf0a3b", "task_id": "vn_S_NARR_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.569639682769775, "target_z": 2.2031250778678024, "ce": 6.3641486167907715, "genuineness": 4.362998008728027, "identity": 0.43414896726608276, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/de/7513ac99a550bb7a957223cb.mp3" }, { "key": "original-vn_S_NARR_reg-de-1-seed1702-scale0.75", "bucket": "vn_S_NARR_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "fd20657234c474e10a5f39b16d6dea61f8bc29057cda7ae34f499642a2cf4683", "task_id": "vn_S_NARR_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.15476655960083, "target_z": 1.914062518169154, "ce": 6.252951145172119, "genuineness": 4.068024158477783, "identity": 0.545806884765625, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/de/ea70eaed792785bc6394dceb.mp3" }, { "key": "original-vn_S_NARR_reg-de-1-seed1702-scale1.0", "bucket": "vn_S_NARR_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "885e2bdc6bbf2b0a3dfb8e4f9d771a9a07cecb62a5bbadaed49b9c0b978229a0", "task_id": "vn_S_NARR_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5247883796691895, "target_z": 2.171874963661692, "ce": 6.41180419921875, "genuineness": 4.2751336097717285, "identity": 0.3988301157951355, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/de/960a614c03a184c4c909d9a5.mp3" }, { "key": "original-vn_S_NARR_reg-en-0-seed1701-scale0.0", "bucket": "vn_S_NARR_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "f3db8793773f18cb1898ae3de44de76f5ce7188616b10904ce1c0db15f0a3cac", "task_id": "vn_S_NARR_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3474459648132324, "target_z": 1.3515624558749118, "ce": 6.284721851348877, "genuineness": 4.042920112609863, "identity": 0.33742785453796387, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/en/b9fbe4465d9a3ec807a2aaf3.mp3" }, { "key": "original-vn_S_NARR_reg-en-0-seed1701-scale0.5", "bucket": "vn_S_NARR_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "4a09b906be714e4e8c6329f2621a8f125044d17afdf5221b9a2c24bca5168bee", "task_id": "vn_S_NARR_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5380632877349854, "target_z": 1.4843749428969448, "ce": 6.268836498260498, "genuineness": 3.911123275756836, "identity": 0.24555402994155884, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/en/8b93e031d9b33480106c2a11.mp3" }, { "key": "original-vn_S_NARR_reg-en-0-seed1701-scale0.75", "bucket": "vn_S_NARR_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "027d6a28098c1300df84f1abe199d59e7ebacee50b71c1c5d3fb7d90f3ac617e", "task_id": "vn_S_NARR_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3474459648132324, "target_z": 1.3515624558749118, "ce": 6.125868320465088, "genuineness": 3.8671910762786865, "identity": 0.2386413812637329, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/en/4afc465b13e21fcad8cfb5dd.mp3" }, { "key": "original-vn_S_NARR_reg-en-0-seed1701-scale1.0", "bucket": "vn_S_NARR_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "3403d4ead08b3040a3d5730b50ae00f0d5f55762ea93a8744313be859c2601da", "task_id": "vn_S_NARR_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3158698081970215, "target_z": 0.6328124870220329, "ce": 6.260893821716309, "genuineness": 4.3881025314331055, "identity": 0.29004886746406555, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/en/71af7ef326708906c07f8d9a.mp3" }, { "key": "original-vn_S_NARR_reg-en-0-seed1702-scale0.0", "bucket": "vn_S_NARR_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "938ce22fa4ed7b0918edae05816435b77216ea56d14419a98ed422e8f61a4402", "task_id": "vn_S_NARR_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8652963638305664, "target_z": 1.0156250571030552, "ce": 6.260893821716309, "genuineness": 4.300237655639648, "identity": 0.2827536165714264, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/en/570e9f17a07ff8379d0d8610.mp3" }, { "key": "original-vn_S_NARR_reg-en-0-seed1702-scale0.5", "bucket": "vn_S_NARR_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "63d9a5ed09b42d8c2174f7bdef06a08eb7ffe730bbb68d54bc83720e041b1226", "task_id": "vn_S_NARR_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.145616054534912, "target_z": 1.2109376064193302, "ce": 6.197352409362793, "genuineness": 4.049196243286133, "identity": 0.33785712718963623, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/en/17df5baff766e78da280adba.mp3" }, { "key": "original-vn_S_NARR_reg-en-0-seed1702-scale0.75", "bucket": "vn_S_NARR_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "635f13a7616b18c670b1e7fab7090862663c2ef3d41d0937dea2b4d153e23f94", "task_id": "vn_S_NARR_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4932122230529785, "target_z": 1.4531249948088132, "ce": 6.09409761428833, "genuineness": 3.9173994064331055, "identity": 0.2699850797653198, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/en/fae00ece65fdca336a41eda6.mp3" }, { "key": "original-vn_S_NARR_reg-en-0-seed1702-scale1.0", "bucket": "vn_S_NARR_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "421ac3802745825f64801cbcb65d546a7dbd58d5b37593f6cadfe9a4a6ed4476", "task_id": "vn_S_NARR_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0671262741088867, "target_z": 1.156249906558637, "ce": 6.2767791748046875, "genuineness": 4.218649387359619, "identity": 0.2782273292541504, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/en/ecf2715b88a2f5679914a789.mp3" }, { "key": "original-vn_S_NARR_reg-en-1-seed1701-scale0.0", "bucket": "vn_S_NARR_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "b8e18fdc15100633cce72b3a1780c55fd7d6d0b8bc496ac89275156fccb6151e", "task_id": "vn_S_NARR_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.336233139038086, "target_z": 1.3437499273233842, "ce": 6.252951145172119, "genuineness": 4.193545341491699, "identity": 0.3196612298488617, "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_S_NARR_reg/en/d7f8e4b2de6d2cc43d264d86.mp3" }, { "key": "original-vn_S_NARR_reg-en-1-seed1701-scale0.5", "bucket": "vn_S_NARR_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "687a314ff8a5c5cd876eb4c4d62d1a4074beaed0764b1ef580ccf41c7e3f6d28", "task_id": "vn_S_NARR_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.695042371749878, "target_z": 1.5937500103823736, "ce": 6.538887023925781, "genuineness": 4.419482707977295, "identity": 0.43317216634750366, "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_S_NARR_reg/en/8f0288ca1ebe045dd3eefc4a.mp3" }, { "key": "original-vn_S_NARR_reg-en-1-seed1701-scale0.75", "bucket": "vn_S_NARR_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "0015293810a066a5f2c240bdab8594eee3b95849098098011ee0e6830ab656f1", "task_id": "vn_S_NARR_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.459573984146118, "target_z": 1.4296875752722091, "ce": 6.340320587158203, "genuineness": 4.519899368286133, "identity": 0.40945708751678467, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/en/40a698d57c72044c9abeda88.mp3" }, { "key": "original-vn_S_NARR_reg-en-1-seed1701-scale1.0", "bucket": "vn_S_NARR_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "0f52d29eac7c924e6c3073c2633a8ab28846c099a6ba89514537f46e7c4a6034", "task_id": "vn_S_NARR_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7286806106567383, "target_z": 1.6171874299189777, "ce": 6.435632705688477, "genuineness": 4.570107460021973, "identity": 0.41247081756591797, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/en/33feafe02b18479c4d9313fb.mp3" }, { "key": "original-vn_S_NARR_reg-en-1-seed1702-scale0.0", "bucket": "vn_S_NARR_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "1d2e309b1856d642489801f841414acaa619540c75138a4d890e59485494063b", "task_id": "vn_S_NARR_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5941271781921387, "target_z": 1.5234374195366038, "ce": 6.133810997009277, "genuineness": 4.193545341491699, "identity": 0.3600132167339325, "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_S_NARR_reg/en/0a11b96db916f8f85666a426.mp3" }, { "key": "original-vn_S_NARR_reg-en-1-seed1702-scale0.5", "bucket": "vn_S_NARR_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "c718757f8effa4cad795afacf697479e7a596de9da0a7cc926ea9fb78da3cfec", "task_id": "vn_S_NARR_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.042638778686523, "target_z": 1.8359375648898355, "ce": 6.395919322967529, "genuineness": 4.444586753845215, "identity": 0.350911021232605, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/en/4c1c104bba836f3e8cd82efc.mp3" }, { "key": "original-vn_S_NARR_reg-en-1-seed1702-scale0.75", "bucket": "vn_S_NARR_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "79ee649fd6e10140cfe14b4c4e0b7a81f4d67815ae2d4ad6ff1c80caeab23937", "task_id": "vn_S_NARR_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6389784812927246, "target_z": 1.5546875337427144, "ce": 6.260893821716309, "genuineness": 4.444586753845215, "identity": 0.4262235760688782, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/en/673a54efa880612d7f659d6a.mp3" }, { "key": "original-vn_S_NARR_reg-en-1-seed1702-scale1.0", "bucket": "vn_S_NARR_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "01633f6192cd607313aa67867be8d0c640f0a5cee81362434713b3fdc1813f21", "task_id": "vn_S_NARR_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.627765655517578, "target_z": 1.5468750051911868, "ce": 6.324435234069824, "genuineness": 4.287685871124268, "identity": 0.37024641036987305, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/en/3cab9e19d506da7d98998372.mp3" }, { "key": "original-vn_S_NARR_reg-es-0-seed1701-scale0.0", "bucket": "vn_S_NARR_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "bb44d5d7646d06bfb004b22dac81fbaa441c17bf3ce5629585af2de656f4e8ce", "task_id": "vn_S_NARR_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25568151473999, "target_z": 1.9843749428969448, "ce": 6.260893821716309, "genuineness": 4.243753433227539, "identity": 0.42264848947525024, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/es/c66b9d8129674a05d694b6dd.mp3" }, { "key": "original-vn_S_NARR_reg-es-0-seed1701-scale0.5", "bucket": "vn_S_NARR_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "c4b3d2492171638e2408632aaf1fdf34eaae9fc60e3a53fed8d4c3a35373edb4", "task_id": "vn_S_NARR_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.39023494720459, "target_z": 2.0781249532793185, "ce": 5.990842819213867, "genuineness": 4.099404335021973, "identity": 0.6863155364990234, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/es/f1efbc1887874256a82559fa.mp3" }, { "key": "original-vn_S_NARR_reg-es-0-seed1701-scale0.75", "bucket": "vn_S_NARR_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "a44a943de2ea151a02b3574a956918a7f04ba3558b0bfefad55611e903c9dd76", "task_id": "vn_S_NARR_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.569639682769775, "target_z": 2.2031250778678024, "ce": 6.006728172302246, "genuineness": 4.13078498840332, "identity": 0.7870410680770874, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/es/d26c57a4bfe018ce698cf174.mp3" }, { "key": "original-vn_S_NARR_reg-es-0-seed1701-scale1.0", "bucket": "vn_S_NARR_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "bbe40cd8fbed453b705e9bdae87b1734168322d41fcb388f6bf116f910600c45", "task_id": "vn_S_NARR_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.547214031219482, "target_z": 2.1875000207647473, "ce": 5.967014789581299, "genuineness": 4.042920112609863, "identity": 0.6136136651039124, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/es/b21c1ae9cebe4504e9b9ab78.mp3" }, { "key": "original-vn_S_NARR_reg-es-0-seed1702-scale0.0", "bucket": "vn_S_NARR_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "b3a4ca199c9d592516cd26fed02694f8116c2bfbc890822dbd132b2806c7a1e1", "task_id": "vn_S_NARR_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.502363204956055, "target_z": 2.1562502387945948, "ce": 6.086154937744141, "genuineness": 4.20609712600708, "identity": 0.7254549264907837, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/es/1841082651f5486e1b546550.mp3" }, { "key": "original-vn_S_NARR_reg-es-0-seed1702-scale0.5", "bucket": "vn_S_NARR_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "bef432cab4b0f6c9fde36f889dd0b1a2a91ed4ab0f013005cc4192139f191a4c", "task_id": "vn_S_NARR_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.009000301361084, "target_z": 1.8124999792352525, "ce": 5.9431867599487305, "genuineness": 4.256305694580078, "identity": 0.801357626914978, "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_S_NARR_reg/es/4f2cec6db4d375bbc6c19835.mp3" }, { "key": "original-vn_S_NARR_reg-es-0-seed1702-scale0.75", "bucket": "vn_S_NARR_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "f18092482e7079e414b3e8c6261c07b7c52c2a036893c3625846cb1007be5872", "task_id": "vn_S_NARR_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.143553733825684, "target_z": 1.9062499896176264, "ce": 5.903473377227783, "genuineness": 3.8044304847717285, "identity": 0.6454803943634033, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/es/127127f337b72f976c88572d.mp3" }, { "key": "original-vn_S_NARR_reg-es-0-seed1702-scale1.0", "bucket": "vn_S_NARR_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "6b958979ee6f85b4b15b464d9473d0529f923a295e09669ad62badc9ccd3427b", "task_id": "vn_S_NARR_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.121128082275391, "target_z": 1.890624932514571, "ce": 5.919358730316162, "genuineness": 4.137060642242432, "identity": 0.5771486759185791, "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_S_NARR_reg/es/49de34a29d03d43f07c18488.mp3" }, { "key": "original-vn_S_NARR_reg-es-1-seed1701-scale0.0", "bucket": "vn_S_NARR_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "96597b76e68d8b1431a0a9e7ea458f14b824f04fd458e9f3fb67b0c9960ba5e0", "task_id": "vn_S_NARR_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.479937553405762, "target_z": 2.1406251816915396, "ce": 6.260893821716309, "genuineness": 4.457139015197754, "identity": 0.779141902923584, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/es/f802caefc379534d6e95065a.mp3" }, { "key": "original-vn_S_NARR_reg-es-1-seed1701-scale0.5", "bucket": "vn_S_NARR_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "eb6d3d50b34d4908dc9533e0c7b44e7c60e9c15d7d39f1d2e01405d59d304129", "task_id": "vn_S_NARR_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.25568151473999, "target_z": 1.9843749428969448, "ce": 6.157639026641846, "genuineness": 4.469690799713135, "identity": 0.6205154061317444, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/es/c039813393554218da45aace.mp3" }, { "key": "original-vn_S_NARR_reg-es-1-seed1701-scale0.75", "bucket": "vn_S_NARR_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "60037b6b7368eb3581e69b1e24cd66b5ccea3a5cb2f68ea435422033952b7bd7", "task_id": "vn_S_NARR_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.941723585128784, "target_z": 1.7656249740440657, "ce": 6.038498878479004, "genuineness": 4.6956281661987305, "identity": 0.7515430450439453, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/es/ab8933fd3331039a4e28c9c2.mp3" }, { "key": "original-vn_S_NARR_reg-es-1-seed1701-scale1.0", "bucket": "vn_S_NARR_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "de0d9a59b48e1114358e7dcf0a8de253da46d9501410d5b02dbb4623a5d39dbb", "task_id": "vn_S_NARR_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.15476655960083, "target_z": 1.914062518169154, "ce": 6.300607204437256, "genuineness": 4.6705241203308105, "identity": 0.7058498859405518, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/es/045b033a954e324608f88e26.mp3" }, { "key": "original-vn_S_NARR_reg-es-1-seed1702-scale0.0", "bucket": "vn_S_NARR_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "c2833e45efd697cbbec11395843b7e5933749a9571ff9cb3efad0cb607501d91", "task_id": "vn_S_NARR_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.173524379730225, "genuineness": 4.532451152801514, "identity": 0.6260011196136475, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/es/04850208697ae9afb164e50d.mp3" }, { "key": "original-vn_S_NARR_reg-es-1-seed1702-scale0.5", "bucket": "vn_S_NARR_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "aac4baee65d3071c7ad3fc376e93951a71613479db91fa297cc44e6f77c201a7", "task_id": "vn_S_NARR_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.986574649810791, "target_z": 1.7968749221321973, "ce": 6.205295085906982, "genuineness": 4.331617832183838, "identity": 0.6775878667831421, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/es/ee7edd6a3c229554431ad057.mp3" }, { "key": "original-vn_S_NARR_reg-es-1-seed1702-scale0.75", "bucket": "vn_S_NARR_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "8365fe215af0caa55cafdab46709f8a7078290d04d676844727aa4aeb290aa90", "task_id": "vn_S_NARR_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165979385375977, "target_z": 1.9218750467206815, "ce": 6.09409761428833, "genuineness": 4.369274139404297, "identity": 0.5830948948860168, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/es/339e5267dfe6f1f7163cd44f.mp3" }, { "key": "original-vn_S_NARR_reg-es-1-seed1702-scale1.0", "bucket": "vn_S_NARR_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "ae723c11ece7bce912d1c91a614a2c7c2d03e9ee849e896f3bbfced77c4e5da1", "task_id": "vn_S_NARR_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.726618766784668, "target_z": 2.3125001453532317, "ce": 6.141753673553467, "genuineness": 4.21237325668335, "identity": 0.6490904688835144, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/es/76541feb5b57ce403ad1ac02.mp3" }, { "key": "original-vn_S_NARR_reg-fr-0-seed1701-scale0.0", "bucket": "vn_S_NARR_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "c36a5f43ee915e2a4275985dc22f5f303bd8f14377d1bf5b23a2d1e3c62e1867", "task_id": "vn_S_NARR_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.313807725906372, "target_z": 1.328125036338308, "ce": 6.046441555023193, "genuineness": 4.362998008728027, "identity": 0.5709632635116577, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/fr/2e965b2af3920159eccd0050.mp3" }, { "key": "original-vn_S_NARR_reg-fr-0-seed1701-scale0.5", "bucket": "vn_S_NARR_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "0219812d602a3b27f3af4944c02e45389b4fd86500eec5f05fa9f6ded6c27e1d", "task_id": "vn_S_NARR_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7062551975250244, "target_z": 1.6015625389339012, "ce": 6.149696350097656, "genuineness": 4.369274139404297, "identity": 0.8109839558601379, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/fr/72c91ae2e418ae23e140e3d7.mp3" }, { "key": "original-vn_S_NARR_reg-fr-0-seed1701-scale0.75", "bucket": "vn_S_NARR_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "0a39aead4d3f6d997a5a416f753ae4412f4db00b5a280e98a9d88aff41045447", "task_id": "vn_S_NARR_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8295958042144775, "target_z": 1.6875000207647475, "ce": 6.1020402908325195, "genuineness": 4.18726921081543, "identity": 0.7419781684875488, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/fr/f2318a19c94d6fe90b870ea7.mp3" }, { "key": "original-vn_S_NARR_reg-fr-0-seed1701-scale1.0", "bucket": "vn_S_NARR_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "6aa1724c6d315cb8f0b14ba85b7404a4ca6de3d71218bfd4c96b2c803b3bb1eb", "task_id": "vn_S_NARR_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.717468023300171, "target_z": 1.609375067485429, "ce": 6.046441555023193, "genuineness": 4.20609712600708, "identity": 0.6533339023590088, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/fr/e8f0c19adaa7abef05e59ecc.mp3" }, { "key": "original-vn_S_NARR_reg-fr-0-seed1702-scale0.0", "bucket": "vn_S_NARR_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "5a96ae91bba0a602b6d75bbfdb3d93a02ea24260b82e07731915b03c19aceeb1", "task_id": "vn_S_NARR_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.650191307067871, "target_z": 1.5625000622942422, "ce": 6.09409761428833, "genuineness": 4.3504462242126465, "identity": 0.785273551940918, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/fr/bd53a88f17ac6648fdbfb9b8.mp3" }, { "key": "original-vn_S_NARR_reg-fr-0-seed1702-scale0.5", "bucket": "vn_S_NARR_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "7572f5d4abde95d79b7bd6b2cdc985c3d4876e9cdba955adb445fd31a2d838b3", "task_id": "vn_S_NARR_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.650191307067871, "target_z": 1.5625000622942422, "ce": 5.967014789581299, "genuineness": 4.413206577301025, "identity": 0.7791159152984619, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/fr/de02e98b8bfb8ad23b676cb3.mp3" }, { "key": "original-vn_S_NARR_reg-fr-0-seed1702-scale0.75", "bucket": "vn_S_NARR_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "93da75e25ea309a4b3b905b11a624bc38f742e56b12e4247bee0065740a13c01", "task_id": "vn_S_NARR_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6165528297424316, "target_z": 1.5390624766396592, "ce": 6.078212261199951, "genuineness": 4.369274139404297, "identity": 0.5927785634994507, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/fr/337482aaecdc2312d779c296.mp3" }, { "key": "original-vn_S_NARR_reg-fr-0-seed1702-scale1.0", "bucket": "vn_S_NARR_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "b42228913c6ea46d1f46cbca4ded0ac35417b0a37a7dded96cc28f429007a0ee", "task_id": "vn_S_NARR_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8744468688964844, "target_z": 1.718749968852879, "ce": 6.054384231567383, "genuineness": 4.049196243286133, "identity": 0.6885862946510315, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/fr/3b24e55a69e000d0d3141447.mp3" }, { "key": "original-vn_S_NARR_reg-fr-1-seed1701-scale0.0", "bucket": "vn_S_NARR_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "3f164a0d0324c6496e87db8f8f9415d12cc4c2fb5ebbe2ef1a98badd2e6f3f98", "task_id": "vn_S_NARR_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.412660598754883, "target_z": 2.0937500103823736, "ce": 6.2767791748046875, "genuineness": 4.507347106933594, "identity": 0.7734615802764893, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/fr/918e0ea495620d489bcf054e.mp3" }, { "key": "original-vn_S_NARR_reg-fr-1-seed1701-scale0.5", "bucket": "vn_S_NARR_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "b4e7e624a4bd2026e4ccde4e1de59a9ae0191e84ba6fdfbcd1d2cbe302ca692b", "task_id": "vn_S_NARR_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5941271781921387, "target_z": 1.5234374195366038, "ce": 6.205295085906982, "genuineness": 4.425758361816406, "identity": 0.7512136101722717, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/fr/35e8c7d5d29b3f472a66b1c5.mp3" }, { "key": "original-vn_S_NARR_reg-fr-1-seed1701-scale0.75", "bucket": "vn_S_NARR_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "1906200210a830a80d715fded04466de93cccf0452edccff7b18aa156f469970", "task_id": "vn_S_NARR_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0559134483337402, "target_z": 1.148437378007109, "ce": 6.141753673553467, "genuineness": 4.607763767242432, "identity": 0.6518651247024536, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/fr/a563c9d9328ac5c1261b914b.mp3" }, { "key": "original-vn_S_NARR_reg-fr-1-seed1701-scale1.0", "bucket": "vn_S_NARR_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "20c75819a336431f1a4b31e02cfda3d3a01427b7cee31b7b62c9f22e1a2dd77f", "task_id": "vn_S_NARR_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.076276779174805, "target_z": 1.8593748183084604, "ce": 6.157639026641846, "genuineness": 4.331617832183838, "identity": 0.7683711051940918, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/fr/32fdd17acb998a5c8f716a3f.mp3" }, { "key": "original-vn_S_NARR_reg-fr-1-seed1702-scale0.0", "bucket": "vn_S_NARR_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "95e09ea574282b5533ce2fcb6e224db28f556f113a5d2a86ce41cd98511128fc", "task_id": "vn_S_NARR_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4932122230529785, "target_z": 1.4531249948088132, "ce": 6.070269584655762, "genuineness": 4.632867813110352, "identity": 0.7144253849983215, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/fr/e4f272b3859d5d2e036bd517.mp3" }, { "key": "original-vn_S_NARR_reg-fr-1-seed1702-scale0.5", "bucket": "vn_S_NARR_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "b83023b9b69b70ac6223c7312cb70eef80471a8ac3557e1d129aa8d13a5cdfe1", "task_id": "vn_S_NARR_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9529364109039307, "target_z": 1.7734375025955935, "ce": 6.062326908111572, "genuineness": 4.444586753845215, "identity": 0.764962911605835, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/fr/1d472eb0ecc356af188e9581.mp3" }, { "key": "original-vn_S_NARR_reg-fr-1-seed1702-scale0.75", "bucket": "vn_S_NARR_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "41108a8536e1d9e474707eafe3a51f49f62599c40a87476ec398b1734140b1a4", "task_id": "vn_S_NARR_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165979385375977, "target_z": 1.9218750467206815, "ce": 6.348263263702393, "genuineness": 4.68307638168335, "identity": 0.7862907648086548, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/fr/84d5964c2568f75075929d61.mp3" }, { "key": "original-vn_S_NARR_reg-fr-1-seed1702-scale1.0", "bucket": "vn_S_NARR_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "0d5c161f808c0f55e7f06341cbb3fc46b1b8626e2235253883da79838c88ad0c", "task_id": "vn_S_NARR_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8071701526641846, "target_z": 1.671874963661692, "ce": 6.181467056274414, "genuineness": 4.482243061065674, "identity": 0.7461926937103271, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/fr/54d0e593954351f5f4b5f5e6.mp3" }, { "key": "original-vn_S_NARR_reg-nl-0-seed1701-scale0.0", "bucket": "vn_S_NARR_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "7369c74af46908532de8b9bac84184ad4a4051c04493c08e7172fc051837ee6b", "task_id": "vn_S_NARR_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.336233139038086, "target_z": 1.3437499273233842, "ce": 6.141753673553467, "genuineness": 3.9927117824554443, "identity": 0.49149221181869507, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/nl/910c7d1ba4689d9b7cba564b.mp3" }, { "key": "original-vn_S_NARR_reg-nl-0-seed1701-scale0.5", "bucket": "vn_S_NARR_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "36b94e4babebb832a62bb214e91cb3cd5878cd348c3e22741b089362fddf3d8a", "task_id": "vn_S_NARR_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.717468023300171, "target_z": 1.609375067485429, "ce": 6.109982967376709, "genuineness": 4.143336772918701, "identity": 0.651097297668457, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/nl/4a366fd45f148d04dc38e793.mp3" }, { "key": "original-vn_S_NARR_reg-nl-0-seed1701-scale0.75", "bucket": "vn_S_NARR_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "6b9dee48cd3cd7986c141115a3c50729c61cbd4b9319cc96d06ab72dcf3efc27", "task_id": "vn_S_NARR_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3698716163635254, "target_z": 1.3671875129779671, "ce": 6.157639026641846, "genuineness": 4.2249250411987305, "identity": 0.6518561244010925, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/nl/03d59b7c373b5a807d8a08db.mp3" }, { "key": "original-vn_S_NARR_reg-nl-0-seed1701-scale1.0", "bucket": "vn_S_NARR_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "2afbb72c04d3ef315054e5d1fd3e15e2583d09f9fc3ae34226dae6afc64495fe", "task_id": "vn_S_NARR_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.291382074356079, "target_z": 1.3124999792352525, "ce": 6.165581703186035, "genuineness": 4.162164688110352, "identity": 0.580532431602478, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/nl/c8e5448410200bd6ec97120d.mp3" }, { "key": "original-vn_S_NARR_reg-nl-0-seed1702-scale0.0", "bucket": "vn_S_NARR_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "d6fdf7bc766cfb86409ec6567ce7c21351c1004c779689779a7ffd1d76e673ed", "task_id": "vn_S_NARR_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.650191307067871, "target_z": 1.5625000622942422, "ce": 6.149696350097656, "genuineness": 4.268857479095459, "identity": 0.7850791215896606, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/nl/fb3db7c2b65781233f7e9192.mp3" }, { "key": "original-vn_S_NARR_reg-nl-0-seed1702-scale0.5", "bucket": "vn_S_NARR_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "45260416201a35b212d94595044cfbac40ad6d2fcd63e73c5e7788f8ef84e1a6", "task_id": "vn_S_NARR_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.932572841644287, "target_z": 1.062499896176263, "ce": 6.038498878479004, "genuineness": 4.293961524963379, "identity": 0.6824403405189514, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/nl/7702cea75074225f428ccf7b.mp3" }, { "key": "original-vn_S_NARR_reg-nl-0-seed1702-scale0.75", "bucket": "vn_S_NARR_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "03db89daf7d1f1ec0f1090eca56fb1fd0deb592b14d0f12b42eaf3e252c8adaf", "task_id": "vn_S_NARR_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.02227520942688, "target_z": 1.1249999584705053, "ce": 6.1020402908325195, "genuineness": 4.074300289154053, "identity": 0.5822856426239014, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/nl/831b12fc03606592091c4808.mp3" }, { "key": "original-vn_S_NARR_reg-nl-0-seed1702-scale1.0", "bucket": "vn_S_NARR_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "d5dd8fc11249c8104b7bd2a4b83fcee47d2a3e077499a37ca9c24df641ecd461", "task_id": "vn_S_NARR_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.381084442138672, "target_z": 1.3750000415294947, "ce": 5.919358730316162, "genuineness": 4.218649387359619, "identity": 0.6839962005615234, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/nl/f8048aea3ae65abb002a208b.mp3" }, { "key": "original-vn_S_NARR_reg-nl-1-seed1701-scale0.0", "bucket": "vn_S_NARR_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "9c57cd60d10421db2d29b173f555aca7ff6c155cf231e673adac52f36d7daf9e", "task_id": "vn_S_NARR_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1792542934417725, "target_z": 1.2343750259559343, "ce": 6.2767791748046875, "genuineness": 4.482243061065674, "identity": 0.5941667556762695, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/nl/e4cbff8751ffd1d5c7e63e31.mp3" }, { "key": "original-vn_S_NARR_reg-nl-1-seed1701-scale0.5", "bucket": "vn_S_NARR_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "1dd8162e4417a8e4e83078829cb842cc7cf5700ecc71b95286e9d6d63d529d83", "task_id": "vn_S_NARR_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0202131271362305, "target_z": 1.8203125077867803, "ce": 6.252951145172119, "genuineness": 4.268857479095459, "identity": 0.7793244123458862, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/nl/eda65ef68140f2a4924830b3.mp3" }, { "key": "original-vn_S_NARR_reg-nl-1-seed1701-scale0.75", "bucket": "vn_S_NARR_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "94e6d53fcf019ba9450153d8a47db73cca707e22287f5dae1563fb1f910b2582", "task_id": "vn_S_NARR_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3922972679138184, "target_z": 1.3828125700810223, "ce": 6.125868320465088, "genuineness": 4.18099308013916, "identity": 0.6516532897949219, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/nl/3b034a3c3fa42930de9d00bb.mp3" }, { "key": "original-vn_S_NARR_reg-nl-1-seed1701-scale1.0", "bucket": "vn_S_NARR_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "672b42b5c50233641c0ff17573843d91c00c44544b2d2441580ef802253f5a0a", "task_id": "vn_S_NARR_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.85408353805542, "target_z": 1.0078125285515276, "ce": 6.149696350097656, "genuineness": 4.507347106933594, "identity": 0.6975088119506836, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/nl/4e81d65a0d6694e4e0e7de4e.mp3" }, { "key": "original-vn_S_NARR_reg-nl-1-seed1702-scale0.0", "bucket": "vn_S_NARR_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "5df1008d7f4b4920d762ecafc4683845ac8e272ae872f403b00a7da7947a9c16", "task_id": "vn_S_NARR_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.009000301361084, "target_z": 1.8124999792352525, "ce": 6.157639026641846, "genuineness": 4.042920112609863, "identity": 0.6905332803726196, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/nl/6ad5b2bdd0f0d67b50d16a9d.mp3" }, { "key": "original-vn_S_NARR_reg-nl-1-seed1702-scale0.5", "bucket": "vn_S_NARR_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "5687745a432c1817b8301d58d185c45e2f2290498e6d07979811a3e6cf9987df", "task_id": "vn_S_NARR_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.613008499145508, "target_z": 0.8398436630476204, "ce": 5.935244083404541, "genuineness": 4.281409740447998, "identity": 0.5834263563156128, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/nl/6081c68829227c0f76e3727f.mp3" }, { "key": "original-vn_S_NARR_reg-nl-1-seed1702-scale0.75", "bucket": "vn_S_NARR_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "3b3320ec013ec427920e8041850a6dbc0b864f906ca201df48840db92490a166", "task_id": "vn_S_NARR_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.95499849319458, "target_z": 1.0781249532793185, "ce": 6.086154937744141, "genuineness": 4.168440818786621, "identity": 0.5207267999649048, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/nl/03755a74dc0112d1fe810b52.mp3" }, { "key": "original-vn_S_NARR_reg-nl-1-seed1702-scale1.0", "bucket": "vn_S_NARR_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "4c1462fef3d5fd52ba3369a662ef4b5d044f3bcd572014cad39116ef0aa2f1be", "mp3_sha256": "c639a157f5c71a79b586b149e81d5eee88ec5b2db525176da91fb44b0ac67e9d", "task_id": "vn_S_NARR_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Narrator Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.143553733825684, "target_z": 1.9062499896176264, "ce": 6.292664527893066, "genuineness": 4.444586753845215, "identity": 0.7623317241668701, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen daar samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_NARR_reg/nl/e8db092a58fd5cb89b3a3ca7.mp3" } ], "source_standard_audio_unchanged": true }