{ "complete": true, "report": { "bucket": "vn_S_CASU_reg", "label": "Strong Casual Style", "definition": { "id": "vn_S_CASU_reg", "category": "talking_style", "head": "vn_S_CASU_reg", "pole": "high", "label": "Strong Casual Style", "floor": 4.0, "canonical_dimension_name": "Casual Style" }, "recipe_status": "direct_manual_recipe", "manual_provenance": [ { "id": "rank4/style_s_casu/full-v2-sealed/sealed-selected", "mission": "style_s_casu", "format_status": "CANONICAL", "heldout": true, "qualified": true, "source_cfg": 3, "source_model": { "label": "Humaneness 1.5 + real-speech rank 4", "repo": "laion/Humaneness-Voice-Large-1.5-RealSpeech-LoRAs", "revision": "fed33e554f094fcdc89fc0124aed2f9d42018714", "rank": 4 }, "metrics": { "ce": 6.287501735687256, "genuineness": 4.532765097618103, "wer": 0.01470995794181278, "identity": 0.6311511784791947, "voiceclap": -0.21498518250882626, "fitness": 0.6065876922348623, "steering_v2": 0.538804825754265, "raw_target": 3.8798326778411867, "target_cdf": 0.538804825754265, "family_hit_all": null, "nominal_hit_all": null, "n": 100, "valid_fraction": 1, "by_arm": { "reference": { "n": 50, "ce": 6.291075954437256, "genuineness": 4.485883092880249, "wer": 0.011323234524847427, "identity": 0.6311511784791947, "fitness": 0.5971404590207908, "steering_v2": 0.5196716561149389 }, "none": { "n": 50, "ce": 6.283927516937256, "genuineness": 4.579647102355957, "wer": 0.018096681358778132, "identity": null, "fitness": 0.6160349254489338, "steering_v2": 0.5579379953935911 } } } } ], "selected_wording": { "id": "manual-e8c5567cec7f", "general": "Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.", "inline": [ "Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.", "Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.", "End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly." ], "cfg": 3, "recipe": "P04", "strategy": "local-Gemma-mutation", "timing_syntax": "canonical", "canonical_labels": true, "neutral_event": "keep", "parent": "cfg-3", "origin": "exact_frozen_manual_recipe" }, "selection": { "manual_no_adapter": { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 0.11252441718981375, "target_raw": 3.894043242931366, "ce": 5.92372727394104, "genuineness": 4.556300234794617, "identity": 0.6006988525390625, "wer": 0.006458333333333333, "valid_fraction": 1, "identity_delta": 0.0037307605147361756, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "manual_adapter": { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": 0.12449342068443718, "target_raw": 3.909314477443695, "ce": 5.9028778076171875, "genuineness": 4.49228458404541, "identity": 0.5892552144825458, "wer": 0.013906088751289989, "valid_fraction": 1, "identity_delta": -0.007712877541780472, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "grid": [ { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 0.11252441718981375, "target_raw": 3.894043242931366, "ce": 5.92372727394104, "genuineness": 4.556300234794617, "identity": 0.6006988525390625, "wer": 0.006458333333333333, "valid_fraction": 1, "identity_delta": 0.0037307605147361756, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": 0.04783934206190351, "target_raw": 3.8115116477012636, "ce": 5.937229871749878, "genuineness": 4.545317125320435, "identity": 0.5891198344528675, "wer": 0.014298245614035088, "valid_fraction": 1, "identity_delta": -0.007848257571458817, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": -0.15305175281016664, "target_raw": 3.555194985866547, "ce": 6.058355617523193, "genuineness": 4.514878344535828, "identity": 0.5477324180305004, "wer": 0.013596491228070174, "valid_fraction": 0.95, "identity_delta": -0.04923567399382591, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.25 }, "n": 20, "target_z": -0.35418701724458335, "target_raw": 3.2985667884349823, "ce": 6.123485541343689, "genuineness": 4.437055492401123, "identity": 0.596756038069725, "wer": 0.0116593567251462, "valid_fraction": 1, "identity_delta": -0.00021205395460128785, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.5 }, "n": 20, "target_z": -0.5104003844524901, "target_raw": 3.0992543756961823, "ce": 6.1131600618362425, "genuineness": 4.422306776046753, "identity": 0.5773192435503006, "wer": 0.005756578947368421, "valid_fraction": 1, "identity_delta": -0.019648848474025725, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.0 }, "n": 20, "target_z": -0.7501953094890196, "target_raw": 2.793300372362137, "ce": 6.140959405899048, "genuineness": 4.397516417503357, "identity": 0.5823082640767098, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.014659827947616577, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 0.26071777171531046, "target_raw": 4.0831229329109195, "ce": 5.755739855766296, "genuineness": 4.580462980270386, "identity": 0.5452585197985172, "wer": 0.037579334365325075, "valid_fraction": 0.95, "identity_delta": -0.051709572225809096, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": -0.2191894516058235, "target_raw": 3.4708099901676177, "ce": 6.075233793258667, "genuineness": 4.522095847129822, "identity": 0.534945746511221, "wer": 0.008534356725146198, "valid_fraction": 1, "identity_delta": -0.06202234551310539, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": 0.12449342068443718, "target_raw": 3.909314477443695, "ce": 5.9028778076171875, "genuineness": 4.49228458404541, "identity": 0.5892552144825458, "wer": 0.013906088751289989, "valid_fraction": 1, "identity_delta": -0.007712877541780472, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": 0.040417488047524555, "target_raw": 3.8020421147346495, "ce": 5.849264788627624, "genuineness": 4.492284655570984, "identity": 0.5564235717058181, "wer": 0.07235638114895081, "valid_fraction": 0.95, "identity_delta": -0.04054452031850815, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.5 }, "n": 20, "target_z": -0.12501220917885458, "target_raw": 3.5909705996513366, "ce": 6.016060876846313, "genuineness": 4.462473440170288, "identity": 0.5689069420099259, "wer": 0.01346749226006192, "valid_fraction": 0.95, "identity_delta": -0.02806115001440048, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": -0.15632932988122128, "target_raw": 3.5510131299495695, "ce": 5.981311702728272, "genuineness": 4.477849721908569, "identity": 0.6395545050501823, "wer": 0.0662452700378397, "valid_fraction": 0.95, "identity_delta": 0.042586413025856015, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": -0.19440920047467009, "target_raw": 3.502427077293396, "ce": 6.033336186408997, "genuineness": 4.503581500053405, "identity": 0.5477780848741531, "wer": 0.01653766769865841, "valid_fraction": 1, "identity_delta": -0.04919000715017319, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.0 }, "n": 20, "target_z": -0.2997131461953751, "target_raw": 3.368069922924042, "ce": 6.048427295684815, "genuineness": 4.497305393218994, "identity": 0.5964943505823612, "wer": 0.019478844169246647, "valid_fraction": 1, "identity_delta": -0.00047374144196510315, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.5 }, "n": 20, "target_z": -0.33067628713522074, "target_raw": 3.328564095497131, "ce": 6.108791637420654, "genuineness": 4.464983820915222, "identity": 0.5732955142855645, "wer": 0.005902777777777778, "valid_fraction": 1, "identity_delta": -0.023672577738761903, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.25 }, "n": 20, "target_z": -0.35715332682488565, "target_raw": 3.294782078266144, "ce": 6.125868344306946, "genuineness": 4.459335494041443, "identity": 0.5943785607814789, "wer": 0.008187134502923975, "valid_fraction": 1, "identity_delta": -0.0025895312428474426, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.5 }, "n": 20, "target_z": -0.4054443267517287, "target_raw": 3.233167660236359, "ce": 6.110975837707519, "genuineness": 4.460276806354523, "identity": 0.610579389333725, "wer": 0.005263157894736842, "valid_fraction": 1, "identity_delta": 0.01361129730939865, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.5 }, "n": 20, "target_z": -0.43325198240906554, "target_raw": 3.1976879119873045, "ce": 6.116734218597412, "genuineness": 4.425758635997772, "identity": 0.6195315927267074, "wer": 0.011228070175438596, "valid_fraction": 1, "identity_delta": 0.022563500702381133, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.0 }, "n": 20, "target_z": -0.45950929425200177, "target_raw": 3.1641862452030183, "ce": 6.173127245903015, "genuineness": 4.4769083023071286, "identity": 0.5634433805942536, "wer": 0.014705882352941176, "valid_fraction": 0.95, "identity_delta": -0.03352471143007278, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.25 }, "n": 20, "target_z": -0.46083983080278423, "target_raw": 3.162488615512848, "ce": 6.138576626777649, "genuineness": 4.476280760765076, "identity": 0.5570772543549538, "wer": 0.005572755417956656, "valid_fraction": 1, "identity_delta": -0.039890837669372556, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.0 }, "n": 20, "target_z": -0.5277831905838104, "target_raw": 3.0770756781101225, "ce": 6.1782899618148805, "genuineness": 4.449293756484986, "identity": 0.5761785194277763, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.02078957259654999, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.0 }, "n": 20, "target_z": -0.5626098604196781, "target_raw": 3.0326403796672823, "ce": 6.144533586502075, "genuineness": 4.375864028930664, "identity": 0.5771041929721832, "wer": 0.003125, "valid_fraction": 1, "identity_delta": -0.019863899052143096, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 0.26735840645993797, "target_raw": 4.091595709323883, "ce": 5.824443984031677, "genuineness": 4.5610072135925295, "identity": 0.5354227613657713, "wer": 0.03729424664602683, "valid_fraction": 1, "identity_delta": -0.06154533065855503, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 0.24339601381405404, "target_raw": 4.0610221266746525, "ce": 5.7382659912109375, "genuineness": 4.538727283477783, "identity": 0.5462718417868018, "wer": 0.025689714482284142, "valid_fraction": 1, "identity_delta": -0.05069625023752451, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 4.0 }, "n": 20, "target_z": 0.10683592554732017, "target_raw": 3.8867853045463563, "ce": 5.81967830657959, "genuineness": 4.519271564483643, "identity": 0.5333611384034157, "wer": 0.016847265221878224, "valid_fraction": 1, "identity_delta": -0.06360695362091065, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 0.10642091156948048, "target_raw": 3.8862557888031004, "ce": 5.746605801582336, "genuineness": 4.518330121040345, "identity": 0.5386883191764354, "wer": 0.025134158926728586, "valid_fraction": 0.95, "identity_delta": -0.058279772847890855, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 0.06210937051090193, "target_raw": 3.82971875667572, "ce": 5.902282094955444, "genuineness": 4.50797472000122, "identity": 0.5381020694971085, "wer": 0.026701066391468867, "valid_fraction": 1, "identity_delta": -0.058866022527217864, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 3.0 }, "n": 20, "target_z": -0.033312974580148605, "target_raw": 3.7079695224761964, "ce": 5.965624976158142, "genuineness": 4.517702579498291, "identity": 0.5347515061497689, "wer": 0.011721491228070174, "valid_fraction": 1, "identity_delta": -0.062216585874557494, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 3.0 }, "n": 20, "target_z": -0.05375977773735302, "target_raw": 3.681881475448608, "ce": 5.957880806922913, "genuineness": 4.500443530082703, "identity": 0.5424081452190876, "wer": 0.011601307189542484, "valid_fraction": 1, "identity_delta": -0.05455994680523872, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": -0.21342162249078833, "target_raw": 3.47816915512085, "ce": 6.085360670089722, "genuineness": 4.473142671585083, "identity": 0.5344767861068249, "wer": 0.006274509803921568, "valid_fraction": 1, "identity_delta": -0.06249130591750145, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.0 }, "n": 20, "target_z": -0.2135376085220115, "target_raw": 3.478021168708801, "ce": 6.10462167263031, "genuineness": 4.511426544189453, "identity": 0.5410011865198612, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.055966905504465106, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.25 }, "n": 20, "target_z": -0.48341063866782186, "target_raw": 3.1336905539035795, "ce": 6.1481078386306764, "genuineness": 4.47973256111145, "identity": 0.5216311305761338, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": -0.0753369614481926, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" } ], "development_standard": [ { "scale": 0.0, "n": 20, "target_z": -1.053710936350353, "target_raw": 2.406045216321945, "ce": 6.180275654792785, "genuineness": 4.34354248046875, "identity": 0.5969680920243263, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.5, "n": 20, "target_z": -0.9567383035038824, "target_raw": 2.529772460460663, "ce": 6.13778235912323, "genuineness": 4.355153155326843, "identity": 0.5114360883831978, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.08553200364112853, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.75, "n": 20, "target_z": -0.8140380610581338, "target_raw": 2.7118434965610505, "ce": 6.179084205627442, "genuineness": 4.383709132671356, "identity": 0.5223133660852909, "wer": 0.005263157894736842, "valid_fraction": 1, "identity_delta": -0.07465472593903541, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "scale": 1.0, "n": 20, "target_z": -1.067089837226209, "target_raw": 2.3889750957489015, "ce": 6.147710680961609, "genuineness": 4.249715673923492, "identity": 0.5518658511340618, "wer": 0.00804093567251462, "valid_fraction": 1, "identity_delta": -0.04510224089026451, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ] }, "heldout": { "standard": { "n": 50, "target_z": -0.7763769473138078, "target_raw": 2.7598952579498293, "ce": 6.239607410430908, "genuineness": 4.378813800811767, "identity": 0.5984692323207855, "wer": 0.010144927536231883, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "manual": { "n": 50, "target_z": 0.030966782904963304, "target_raw": 3.7899839735031127, "ce": 5.987586421966553, "genuineness": 4.533329925537109, "identity": 0.6220511162281036, "wer": 0.027190055177469363, "valid_fraction": 0.98, "identity_delta": 0.023581883907318114, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "adapter": { "n": 50, "target_z": 0.07578857654381994, "target_raw": 3.8471720361709596, "ce": 5.962408142089844, "genuineness": 4.535338220596313, "identity": 0.579516797363758, "wer": 0.027015433841520796, "valid_fraction": 1, "identity_delta": -0.018952434957027434, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "matched_no_adapter": { "n": 50, "target_z": 0.030966782904963304, "target_raw": 3.7899839735031127, "ce": 5.987586421966553, "genuineness": 4.533329925537109, "identity": 0.6220511162281036, "wer": 0.027190055177469363, "valid_fraction": 0.98, "identity_delta": 0.023581883907318114, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" } }, "adapter_vs_matched_cfg": { "n": 50, "delta_z": 0.04482179363885664, "ci95_z": [ -0.03737792713136011, 0.14386721944558975 ], "language_deltas": { "en": 0.2255127411318663, "de": -0.014770457627212083, "fr": 0.05428469534700182, "nl": 0.057568331459561896, "es": -0.09848634211693473 }, "languages_improved": 3, "clear_gain": false, "delta_ce": -0.025178279876708985, "delta_genuineness": 0.0020082950592041015, "delta_wer": -0.00017462133594856697, "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": 0.04482179363885664, "ci95_z": [ -0.03737792713136011, 0.14386721944558975 ], "language_deltas": { "en": 0.2255127411318663, "de": -0.014770457627212083, "fr": 0.05428469534700182, "nl": 0.057568331459561896, "es": -0.09848634211693473 }, "languages_improved": 3, "clear_gain": false, "delta_ce": -0.025178279876708985, "delta_genuineness": 0.0020082950592041015, "delta_wer": -0.00017462133594856697, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "manual_vs_standard": { "n": 50, "delta_z": 0.8073437302187711, "ci95_z": [ 0.7076561973717121, 0.9063964810656587 ], "language_deltas": { "en": 0.6276854751087381, "de": 0.7362304427734508, "fr": 0.8136230610950393, "nl": 0.9401855533577321, "es": 0.9189941187588953 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.25202098846435544, "delta_genuineness": 0.1545161247253418, "delta_wer": 0.01704512764123748, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "route": { "adapter": false, "scale": 0.0, "cfg": 4.0, "reason": "Use manual-only configuration; no clear qualifying incremental adapter gain" }, "reference_only": true, "updated": 1793024388.3942852, "example": { "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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.", "language": "de", "seed": 1701, "scale": 0.0, "cfg": 4.0, "frames": 128, "reference": true } }, "rows": [ { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "0096d00212f603fed821332284149c7569368c1d0a1b73e4a5e747da499ce5cd", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "21039fe2163a90146b983d6614d5fc685570a5f807306e35e700acaab8906d2c", "generation_seconds_batch": 22.683037291979417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.04079008102417, "target_z": 0.2275389182466657, "ce": 5.919358730316162, "genuineness": 4.444586753845215, "identity": 0.6925682425498962, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0096d00212f603fed821332284149c7569368c1d0a1b73e4a5e747da499ce5cd.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": "018e360ff8daab60237771b1f3c7f656daf77a2cae52cb3323ff642aa17c69b9", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2bdba1e62d9754b79a0e2b9d3a1b94eb6daa3df459f733562804368859ba63d4", "generation_seconds_batch": 17.278336070943624, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.702502727508545, "target_z": -0.0375976372260783, "ce": 5.998785495758057, "genuineness": 4.582659721374512, "identity": 0.547799289226532, "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_CASU_reg/018e360ff8daab60237771b1f3c7f656daf77a2cae52cb3323ff642aa17c69b9.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "01a60cdeff8382ceccf9d932d93fefd5bc5ffd8bbb5d6a32a1f51c8ec899153c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "b162997597fba0b5fe9eaa63a3bd48fe7d1b19535488e5144b7ac69dd7070e57", "generation_seconds_batch": 16.076712407986633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6682379245758057, "target_z": -0.06445306569281005, "ce": 6.213237762451172, "genuineness": 4.883909225463867, "identity": 0.8046466112136841, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/01a60cdeff8382ceccf9d932d93fefd5bc5ffd8bbb5d6a32a1f51c8ec899153c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "01c173a0ce6684104fd616fd26db931ec3b57a61b9ff48db2a9a660aeb86526b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "e215060a473b17bda9cb3d0da5c1cc157a8966334b6c98696fa5cc8a05e9638e", "generation_seconds_batch": 22.850207146955654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1057605743408203, "target_z": -1.2890624350358166, "ce": 6.0146708488464355, "genuineness": 4.250029563903809, "identity": 0.6886022090911865, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/01c173a0ce6684104fd616fd26db931ec3b57a61b9ff48db2a9a660aeb86526b.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "1db440f259086b58ab71268b96c0c97fffad8bdaba9ed00c7ade81494f38c6ef", "generation_seconds_batch": 14.784412681008689, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3150875568389893, "target_z": -1.125000035036863, "ce": 6.054384231567383, "genuineness": 4.469690799713135, "identity": 0.6090685725212097, "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_CASU_reg/01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "0231db464cab263b49d06109d893fe98c6df9101e0747343fa86e529924d682d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "2dd930db0f7149bfe82898301e293b4c1c5f831734dc8204f87ee2f20a3d3eea", "generation_seconds_batch": 23.316566491033882, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2620434761047363, "target_z": -0.38281257810300706, "ce": 5.657250881195068, "genuineness": 4.607763767242432, "identity": 0.6487427353858948, "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_CASU_reg/0231db464cab263b49d06109d893fe98c6df9101e0747343fa86e529924d682d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "023d0b3f89e622def34578606c8a49caa6b7746991ec4ec90df642bcd26cd25f", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "8c224551ca34fbccdf7d846d7ca7a516eea9cf1cb0fc7f9451b3dccdcba29326", "generation_seconds_batch": 21.631574525032192, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.226443290710449, "target_z": 0.3730467299254892, "ce": 6.300607204437256, "genuineness": 4.557555198669434, "identity": 0.23867720365524292, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/023d0b3f89e622def34578606c8a49caa6b7746991ec4ec90df642bcd26cd25f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "02fd68b1e6596f1064dd4055a44705adc8703da3fe85676a0e4b9ead8792b01e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "41545b9419d9b385f4bbc1bfc076797f93f06567f43e343cc483f7f4678ee928", "generation_seconds_batch": 23.96699068392627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0626845359802246, "target_z": -0.5390624116779079, "ce": 6.157639026641846, "genuineness": 4.143336772918701, "identity": 0.33050844073295593, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/02fd68b1e6596f1064dd4055a44705adc8703da3fe85676a0e4b9ead8792b01e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "0314472b0b51aa1dcb5c8c28081ad0fad7a6a0a61c2fd727745dc84b24736a16", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "03da7db0710aa26204a0622406ffd955838426802f2bd155c94278c774068feb", "generation_seconds_batch": 16.795615051989444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.119287967681885, "target_z": 0.2890625284674512, "ce": 5.796247482299805, "genuineness": 4.507347106933594, "identity": 0.29690107703208923, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0314472b0b51aa1dcb5c8c28081ad0fad7a6a0a61c2fd727745dc84b24736a16.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "0383d3fde911abadac46bdaed542e932e0c677997961645865dcd5457b914ef1", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "5bd73496d853d66c6fc43cc2533a57cedc55ca6e427d8b36961e36013e0f3d16", "generation_seconds_batch": 23.66319642099552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0046563148498535, "target_z": 0.19921866861228701, "ce": 6.070269584655762, "genuineness": 4.557555198669434, "identity": 0.665023148059845, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Ähm, sag mir, wenn du bereit bist.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0383d3fde911abadac46bdaed542e932e0c677997961645865dcd5457b914ef1.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": "03ec8e696b7a2136c9fe2f2ce2e53ab315867cd1929c0ad624b145a39f5906d3", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2af3af6b1ada151eecb7384c4308686d55ceacbbe0f9fc4cb7e89a8ff105c24f", "generation_seconds_batch": 22.96834774001036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.398390769958496, "target_z": 0.5078125197082354, "ce": 5.903473377227783, "genuineness": 4.532451152801514, "identity": 0.6490034461021423, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Euh dis-moi quand nous pourrons parler.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/03ec8e696b7a2136c9fe2f2ce2e53ab315867cd1929c0ad624b145a39f5906d3.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "0472ca66ce6d6163c360da3f622290a1e275b9859dfd066f504c7623c3cafbcd", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "781b7c954c4d1f22be654cfba5b914525223dabb38e09983995dfa1cc0fe80f6", "generation_seconds_batch": 22.664694448933005, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2171878814697266, "target_z": -0.4179686598530713, "ce": 6.173524379730225, "genuineness": 4.042920112609863, "identity": 0.3197743594646454, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0472ca66ce6d6163c360da3f622290a1e275b9859dfd066f504c7623c3cafbcd.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "04cda0c6c3ee535c9b9fc64e07f4e254261465459682d5de0a3d5c3f5ab7b894", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "623874e5108832b3b64c1d24c6ce2da39c41c60c243494f5d953fb631f248b52", "generation_seconds_batch": 16.846680309972726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3282597064971924, "target_z": -1.8984375708036607, "ce": 6.316492557525635, "genuineness": 4.137060642242432, "identity": 0.5798197388648987, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/04cda0c6c3ee535c9b9fc64e07f4e254261465459682d5de0a3d5c3f5ab7b894.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "04d29486abaf616e158d0ac7ae88df25e88b2cb811a1dc9a82b6544ccaff1383", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "d6ccddcc256d59bef169d401eea867080631223edb333e2cf07360747ed17388", "generation_seconds_batch": 22.68062383902725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.197251796722412, "target_z": -0.43359379270117676, "ce": 5.895530700683594, "genuineness": 4.519899368286133, "identity": 0.7276009917259216, "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_CASU_reg/04d29486abaf616e158d0ac7ae88df25e88b2cb811a1dc9a82b6544ccaff1383.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "04e1184aae32e1c4b2b6a11fac939f758a514020e9d54d31627aa6c2b9fb9d95", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "065a6e24469631393f6cb2cc7ed1cd17c0af8684c47541a2429dde13be851e47", "generation_seconds_batch": 24.664431971963495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.149191856384277, "target_z": 0.31250004087634015, "ce": 5.7684478759765625, "genuineness": 4.65797233581543, "identity": 0.6254016160964966, "hypothesis": "Me he encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/04e1184aae32e1c4b2b6a11fac939f758a514020e9d54d31627aa6c2b9fb9d95.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": "04f8372c613ef13b22c5cdc39a3a0b8e5296130ed3e87261b03b274b8aefa58d", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "178ffc00b125e7fab5259cbe819b4ca282d3d00f1fdf67a95ebd69e7f1363f30", "generation_seconds_batch": 20.660011678934097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5972161293029785, "target_z": -0.12011720437973868, "ce": 5.685050010681152, "genuineness": 4.720732688903809, "identity": 0.833486795425415, "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_CASU_reg/04f8372c613ef13b22c5cdc39a3a0b8e5296130ed3e87261b03b274b8aefa58d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "04fa7db6f3bcb2102354ac644857c952d8cd4cacb8d95185e1d5a1f9cc1930d5", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "5127996a76886e873a06ffd684518f095d8056e1c7481bb543a4bc955d84c728", "generation_seconds_batch": 22.92002897092607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.149191856384277, "target_z": 0.31250004087634015, "ce": 6.0305562019348145, "genuineness": 4.337893962860107, "identity": 0.5759580135345459, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/04fa7db6f3bcb2102354ac644857c952d8cd4cacb8d95185e1d5a1f9cc1930d5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "0500d4e33bc1148f8250068a596a76e2cd177072c6a428ffc929524c71d378b8", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2b08bbe85b3a5a685cf4cfe8756b1cf1a78a45cd3085d95f2abca9c96c0efb42", "generation_seconds_batch": 22.65299829491414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.069447994232178, "target_z": 0.2499998832104567, "ce": 5.855817794799805, "genuineness": 4.796044826507568, "identity": 0.7384136915206909, "hypothesis": "Là, la porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0500d4e33bc1148f8250068a596a76e2cd177072c6a428ffc929524c71d378b8.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": "0560266308104aa2d50c820c47d9f5d3a66613b5924df17491d4d44be801d551", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fbd609865e58cbfe98584caad69c43cfae29fe72614a2a7f577dde9783086ff5", "generation_seconds_batch": 19.861331301974133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.73302960395813, "target_z": -0.01367185109464036, "ce": 5.967014789581299, "genuineness": 4.65797233581543, "identity": 0.7369781732559204, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0560266308104aa2d50c820c47d9f5d3a66613b5924df17491d4d44be801d551.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "05892b6ed427474df5d54249641d685c1d26173f9c8e915e96032484a45998bc", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "8c9fa529f170a7ba23258272ffd63c8291283bc59744d81b2c3a4d0526caad4b", "generation_seconds_batch": 23.715953538077883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.457664728164673, "target_z": -0.2294922000001308, "ce": 6.483288764953613, "genuineness": 4.519899368286133, "identity": 0.2943090796470642, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/05892b6ed427474df5d54249641d685c1d26173f9c8e915e96032484a45998bc.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "a7c05b26950f0afc7b744daed749a928dac3885fc1e7090c0cdbc4cd0410d13c", "generation_seconds_batch": 16.802694446058013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1773159503936768, "target_z": -0.449218738686013, "ce": 6.372091293334961, "genuineness": 4.808597087860107, "identity": 0.7036087512969971, "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_CASU_reg/059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "0607f11caa109cf20dd39a6e14c736fcf16949625fdc76813a0cdd0c0b48ffa0", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b59110bd928ef0f77bb972cfe170a58d535092a0c5d658c8e16ee2987581690f", "generation_seconds_batch": 22.70223337004427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.336091041564941, "target_z": 0.4589844000002616, "ce": 5.669164657592773, "genuineness": 4.7834930419921875, "identity": 0.7949408888816833, "hypothesis": "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_CASU_reg/0607f11caa109cf20dd39a6e14c736fcf16949625fdc76813a0cdd0c0b48ffa0.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "08cb835103c97b1bb79351d6b3ba70935dff81964512ce182cfbc04b6abee3ff", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "554059ed6c86e5ad145efc6cf5956acfee7b2efb6a14345ff10a5d63d1e5486d", "generation_seconds_batch": 15.978954939055257, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0128445625305176, "target_z": -0.5781250569349023, "ce": 6.205295085906982, "genuineness": 4.582659721374512, "identity": 0.6967498064041138, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/08cb835103c97b1bb79351d6b3ba70935dff81964512ce182cfbc04b6abee3ff.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "09b97624e186e32c6a9dfd95d9e1d33e6c60713eb4cc9ef878d82a4520c78942", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "2653e9bff7fab4f14099821c6a46bcbad14ef9592606fb6907b35c466131d853", "generation_seconds_batch": 21.760152954957448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.383439064025879, "target_z": 0.49609395036706017, "ce": 5.454712867736816, "genuineness": 4.118232727050781, "identity": 0.3537815809249878, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/09b97624e186e32c6a9dfd95d9e1d33e6c60713eb4cc9ef878d82a4520c78942.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "0a42295c2c78f23e9d534a9788d4f9929b5638bae8037d1eff22b6062b31f45a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "b4ac6137e44b12f1a292c8168215625c1989abc575e886ef62a6bf56f5e1e20a", "generation_seconds_batch": 24.396074668038636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9560625553131104, "target_z": 0.16113275766365975, "ce": 6.1894097328186035, "genuineness": 4.482243061065674, "identity": 0.5460764169692993, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0a42295c2c78f23e9d534a9788d4f9929b5638bae8037d1eff22b6062b31f45a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "0a4b3adde13da53e711ae5d26b002e782637a4a10a9907fbc05b72ef40a0fc09", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "89f54590cd7dc72dc08ba8313a2ae5ae117d1228078f3983d607601ccc76ad08", "generation_seconds_batch": 23.617501189000905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5162265300750732, "target_z": -0.1835937226274508, "ce": 5.967014789581299, "genuineness": 4.582659721374512, "identity": 0.7850104570388794, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0a4b3adde13da53e711ae5d26b002e782637a4a10a9907fbc05b72ef40a0fc09.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "0a4dcd01edd7db50e96e6c5e2e6b18998529967b8e51daa60f9559972a4e79f2", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "44e15328981992b8d7482ff2e6e820ee0ed1f75e227304989c7b09d82c681b2f", "generation_seconds_batch": 26.318961848970503, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9834744930267334, "target_z": 0.18261717518235285, "ce": 6.078212261199951, "genuineness": 4.494794845581055, "identity": 0.6100187301635742, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0a4dcd01edd7db50e96e6c5e2e6b18998529967b8e51daa60f9559972a4e79f2.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "0ab68433df46847dc61b93d3323f2e1db89e7fef9cdeccb04fc9640507110826", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "ebefd38c8547b507df3ddac5d34db3f31e7a5340bab8c154f64835ad48eff024", "generation_seconds_batch": 22.1925501950318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.242107629776001, "target_z": -0.3984375240878433, "ce": 6.070269584655762, "genuineness": 4.645420074462891, "identity": 0.5662630200386047, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0ab68433df46847dc61b93d3323f2e1db89e7fef9cdeccb04fc9640507110826.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "0aeffc707068b7eeb9a7d815151cc95f16d7aab3aadd97f7efc47beb430c2210", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "9a32a667228d2a0cbafeee8baa11bbc0dda766d99ff014f784066c3c9327f150", "generation_seconds_batch": 22.416251708986238, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.149191856384277, "target_z": 0.31250004087634015, "ce": 6.1020402908325195, "genuineness": 4.394378662109375, "identity": 0.37946271896362305, "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_CASU_reg/0aeffc707068b7eeb9a7d815151cc95f16d7aab3aadd97f7efc47beb430c2210.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "0af433186a2aeef580116502e59a183c9095ed293087284d20edd6a3907c554c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "4ba578a7e6168b869d3741f272a9e003f7c8005ac9ee691b91d4818eb23a0875", "generation_seconds_batch": 23.755018191994168, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.261331558227539, "target_z": 0.40039080584130843, "ce": 6.062326908111572, "genuineness": 4.3881025314331055, "identity": 0.719630777835846, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0af433186a2aeef580116502e59a183c9095ed293087284d20edd6a3907c554c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "0afff29466d8e4ec3ea0c19f4175f88e78a55ac8e7f58c0443031e8dbcb8960d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "00e0b922fcb93b31add4e1d58d4c0dda7f751178985c2bd20665c1724b69cb32", "generation_seconds_batch": 25.366768251988105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.863325357437134, "target_z": -0.695312432116078, "ce": 6.1894097328186035, "genuineness": 4.319066047668457, "identity": 0.3837233781814575, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0afff29466d8e4ec3ea0c19f4175f88e78a55ac8e7f58c0443031e8dbcb8960d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "0bd8fbf7b60c12b6b9b1f2c9ad472cf50f54651c1bc65b77f7e79773fbcdae6c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "45b6fcf018d033d5b50fcc858541b6058071c350ef72c0c5203b3b15a1c92813", "generation_seconds_batch": 16.678186065983027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.658982276916504, "target_z": -0.8554686423346398, "ce": 6.046441555023193, "genuineness": 4.193545341491699, "identity": 0.5977221727371216, "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_CASU_reg/0bd8fbf7b60c12b6b9b1f2c9ad472cf50f54651c1bc65b77f7e79773fbcdae6c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "0c80d1c426592a90c16ed85a9a13a4bbed8d61d8ecd68ebe9f170e207f6f77e0", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "9b8a2463e17aa6f2e87b4588ec14d250122c4d3afd665b20613e01f89b7916c0", "generation_seconds_batch": 22.793955393019132, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3368031978607178, "target_z": -0.32421879708078466, "ce": 6.205295085906982, "genuineness": 4.720732688903809, "identity": 0.6942861080169678, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0c80d1c426592a90c16ed85a9a13a4bbed8d61d8ecd68ebe9f170e207f6f77e0.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "b8c69d14bcac7a0620e30e5af975f6684a3b590d1fd5ec73f229a4f3f140045b", "generation_seconds_batch": 17.05230904999189, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.768629550933838, "target_z": -0.7695313459864059, "ce": 6.165581703186035, "genuineness": 4.331617832183838, "identity": 0.5113776922225952, "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_CASU_reg/0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "0cbc7f708a31c3cde5cf855629346b543c3f6d081988e95856e89a7f8159fd0a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "3ec136b7f7823fc2a37c52c7f80a9ff987635c24cb2b1a2150058b1b6284d51a", "generation_seconds_batch": 22.044438773067668, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.522456407546997, "target_z": -0.17871098540196112, "ce": 6.213237762451172, "genuineness": 4.400654315948486, "identity": 0.44971752166748047, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0cbc7f708a31c3cde5cf855629346b543c3f6d081988e95856e89a7f8159fd0a.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": "0cef88d29d3a6d02268211d199fe029589ade5b21fa3ef176ad8137e2e45a732", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The road turns left behind the bridge.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6278f9a87e8a73e651d25cdb50eb037fee6fa9dfefb29d095fa50751d88ea40e", "generation_seconds_batch": 22.062792692915536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8133962154388428, "target_z": 0.04931639343052279, "ce": 6.356205940246582, "genuineness": 4.469690799713135, "identity": 0.3908769488334656, "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_CASU_reg/0cef88d29d3a6d02268211d199fe029589ade5b21fa3ef176ad8137e2e45a732.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "0d08e25f44b77886713b5cea0e951184d4a6caac813298542a18aa042492a5ee", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "7bf7dc1ef0ce320ee32e99cebedea314496d744a8b90065d7adb381d4a0d20c0", "generation_seconds_batch": 24.726403250941075, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8999929428100586, "target_z": 0.11718756204444487, "ce": 6.038498878479004, "genuineness": 4.770940780639648, "identity": 0.5787255167961121, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0d08e25f44b77886713b5cea0e951184d4a6caac813298542a18aa042492a5ee.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "0d8acda69996388c0de600e27518cf1575e12a3f3c2cbcc1aa4e1247c42886f9", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "04a5e562ee7cb3c21b312ed1cde1d2b7356751b7fc1051562c0d80bd82fbf778", "generation_seconds_batch": 23.918496854952537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.205440044403076, "target_z": -1.2109375182483662, "ce": 6.316492557525635, "genuineness": 4.281409740447998, "identity": 0.5462257862091064, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0d8acda69996388c0de600e27518cf1575e12a3f3c2cbcc1aa4e1247c42886f9.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "0de5c8b77d4852395204605f106d1b154f741ba592d738fa239fcf88793dff51", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "51023044e6edde22d801e4c8f0d7b8b7a3a56841bcfcc3d1655c8664ec5d2346", "generation_seconds_batch": 16.155976719921455, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.924912691116333, "target_z": 0.13671869780967286, "ce": 6.006728172302246, "genuineness": 4.362998008728027, "identity": 0.1860763430595398, "hypothesis": "The door's still open? I put your coat on the chair. Tell me when you are ready.", "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_CASU_reg/0de5c8b77d4852395204605f106d1b154f741ba592d738fa239fcf88793dff51.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "0e08634ec855b4589490145533f3c0a204fd521d57df3021a002e74b5b3b2108", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "22cef712ac2cade5697642df701096183208ad1260e26dd9fc5d60311af5da80", "generation_seconds_batch": 23.322285901056603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.777173042297363, "target_z": 0.8046874277364701, "ce": 6.0146708488464355, "genuineness": 4.482243061065674, "identity": 0.4847610294818878, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0e08634ec855b4589490145533f3c0a204fd521d57df3021a002e74b5b3b2108.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "0e26c83c8dc2b90f2880f2d80062c4bb993944fa1644af18d79591f12a78244a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "e4b2336ab6fcbb51205a9cfb4e40173d2b64ed5f612d3f53ad6c09f053bf9e81", "generation_seconds_batch": 18.02974793803878, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1591596603393555, "target_z": 0.32031242043712366, "ce": 6.062326908111572, "genuineness": 4.68307638168335, "identity": 0.6620824337005615, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi, quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0e26c83c8dc2b90f2880f2d80062c4bb993944fa1644af18d79591f12a78244a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "0e2dbc84ed966e9816acbadb4d42c1f0c9a44edad29d5754e4516ec69817f705", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "cb186eee909ce14d693c398521cb04825ec024b6b2d4fb4489afc50198c8b094", "generation_seconds_batch": 23.012881717993878, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.789722442626953, "target_z": 0.030761805110392747, "ce": 5.824047088623047, "genuineness": 4.6956281661987305, "identity": 0.6895870566368103, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise, dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0e2dbc84ed966e9816acbadb4d42c1f0c9a44edad29d5754e4516ec69817f705.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "0f18001e975362db42e2e13a89ed79e3e748c1673a76d7a201d15df593c9244e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "515b213d502bd94d905201590ecb4e9875f35c5ac5cbe8e0917674b5198fcbad", "generation_seconds_batch": 24.653218284947798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.987924814224243, "target_z": -0.5976561927001304, "ce": 5.855817794799805, "genuineness": 4.3881025314331055, "identity": 0.6657374501228333, "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_CASU_reg/0f18001e975362db42e2e13a89ed79e3e748c1673a76d7a201d15df593c9244e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "0f435e15dcb6003f524e088ce4416387ac8f2c3689e1d0e7758c281523716024", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f9ed8e373b98f76a41298d6ba9f0adeb3ec25a53f4bc9acda5b3c4473e6aa3e1", "generation_seconds_batch": 22.91413032298442, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.593478202819824, "target_z": -0.12304684671503249, "ce": 6.022613525390625, "genuineness": 4.607763767242432, "identity": 0.665097177028656, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0f435e15dcb6003f524e088ce4416387ac8f2c3689e1d0e7758c281523716024.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": "0f4a41b845a82d31d9df8e16bbe6a3b533adca94f344761cc86ad510ac0f3c57", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dce1901469a2cc8eca36aa92b22d7952cfca9c14c4e4537299510b293d480ffd", "generation_seconds_batch": 21.13354949594941, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.2364115715026855, "target_z": 0.3808594832128112, "ce": 5.935244083404541, "genuineness": 4.68307638168335, "identity": 0.682056188583374, "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_CASU_reg/0f4a41b845a82d31d9df8e16bbe6a3b533adca94f344761cc86ad510ac0f3c57.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "0f8968045ab79dea39cc5c940d80fd38fe097c6c68d57f8d4e7efc075a44f527", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "922bd959987f7e65fd6f028e87d190810628747ffbd82ef8f00d1a2553c6d0fb", "generation_seconds_batch": 22.41687322105281, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.963538646697998, "target_z": 0.16699222919751663, "ce": 5.6334228515625, "genuineness": 4.595211505889893, "identity": 0.7704930901527405, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo eh sobre la silla. Dime cuando estás listo.", "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_CASU_reg/0f8968045ab79dea39cc5c940d80fd38fe097c6c68d57f8d4e7efc075a44f527.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "0faa6cbe341055c969bed7d56ff9230dbff7df22999d589f50e027607413a763", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "f9ee9edb1ada03173c60ac3663d01d7896e4cfa894077aeab2c066baac9152c3", "generation_seconds_batch": 22.779580950038508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1556003093719482, "target_z": -1.2499999766420913, "ce": 6.117925643920898, "genuineness": 4.419482707977295, "identity": 0.6262384057044983, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0faa6cbe341055c969bed7d56ff9230dbff7df22999d589f50e027607413a763.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "0fe67a37c428c36cc92883241385852e87c796075cb2e309843417a07756f9ad", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "9a7233c26c2085ddaf6b47ca96b4c2038cad139cdde250edaa6b328b0903c7e6", "generation_seconds_batch": 23.640872180112638, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.500028610229492, "target_z": -0.1962890262769932, "ce": 6.0305562019348145, "genuineness": 4.375550270080566, "identity": 0.5218904614448547, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/0fe67a37c428c36cc92883241385852e87c796075cb2e309843417a07756f9ad.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "103878593a8789be00e99e8ea9524292c284a47598101eab9be95ebf1051146b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "c2a2d28856f983cc56aaaf8246ff19021d6f805a7a26e99cc833bf9e7a130a57", "generation_seconds_batch": 23.75564559700433, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8085014820098877, "target_z": -0.7382812671534642, "ce": 6.141753673553467, "genuineness": 4.545003414154053, "identity": 0.6149731874465942, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/103878593a8789be00e99e8ea9524292c284a47598101eab9be95ebf1051146b.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": "118409d71df55ec718cdfa2ea754b46c143343fbe6882e47237454b07f5d00a5", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f79d63a35333ee11f8ed5340c27aa65cbc4ac28ddccfe8f7d46ec33f89679835", "generation_seconds_batch": 22.663453585933894, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5411462783813477, "target_z": -0.1640625868622228, "ce": 5.863759994506836, "genuineness": 4.6705241203308105, "identity": 0.622342586517334, "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_CASU_reg/118409d71df55ec718cdfa2ea754b46c143343fbe6882e47237454b07f5d00a5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "118fe5dd3c1587e68ba5a0e63d4a5d96598661b37e71733768b9bc62d91a8f24", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "c7e58d046993cb6ac659be970d4cc66fa3786d01872bfcee9b6c425c1a027db3", "generation_seconds_batch": 23.053280350985005, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5536062717437744, "target_z": -0.1542969255479742, "ce": 6.1894097328186035, "genuineness": 4.3881025314331055, "identity": 0.6609687805175781, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/118fe5dd3c1587e68ba5a0e63d4a5d96598661b37e71733768b9bc62d91a8f24.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 127, "seconds_generated": 10.16, "mp3_sha256": "94768e82da5030ab697ebdaad222bf8199c33c6b0fac35bca0836ab09c36e8c6", "generation_seconds_batch": 16.44140391901601, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.255280017852783, "target_z": -1.1718748729913717, "ce": 6.125868320465088, "genuineness": 4.369274139404297, "identity": 0.5977527499198914, "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_CASU_reg/12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "12affd4c30a282c5747fd10e14f9a224f27da31a31b163b2486591c01963eea2", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "b90f751c8041b1f4cf41320eb1fc71495e6b262016535151e745f87a513fb39b", "generation_seconds_batch": 16.980974233942106, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.066956043243408, "target_z": 0.24804678832026084, "ce": 6.006728172302246, "genuineness": 4.582659721374512, "identity": 0.7920567989349365, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/12affd4c30a282c5747fd10e14f9a224f27da31a31b163b2486591c01963eea2.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "130a0f1b757a166b5e14e9e1638860dfe071851e04f4d09cef11be003f7d7f50", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "9e88f8fa5b5955c272c95bd6f2477740690aa4bff96ea76444617ba09eb76944", "generation_seconds_batch": 23.640872180112638, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.064464092254639, "target_z": 0.24609369343006499, "ce": 5.871703147888184, "genuineness": 4.432034492492676, "identity": 0.7805730104446411, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/130a0f1b757a166b5e14e9e1638860dfe071851e04f4d09cef11be003f7d7f50.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": "13248bdb66b8a3042a43dd0cf04b27c19d8884e608276f6e3b1170900c2805fe", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f172b3b2be50af9ddca6d840e4cc6c758373b7a61e49e7f11b2deadbd8617779", "generation_seconds_batch": 23.178023271961138, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.429006814956665, "target_z": -0.25195316496392184, "ce": 6.0305562019348145, "genuineness": 4.2374773025512695, "identity": 0.7619679570198059, "hypothesis": "Las llaves están en la mesa de la cocina. Miren el cajón y luego junto a la ventana, llévatelas cuando te vayas.", "wer": 0.08695652173913043, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/13248bdb66b8a3042a43dd0cf04b27c19d8884e608276f6e3b1170900c2805fe.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "1347a892808ff4331506bca9970ca595d5815f3178fda7f464bbb97dadc17d61", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "33c11dedaef708013676bf21f838b364219cda999f9b5c06134a39e7b34551a8", "generation_seconds_batch": 22.826197736081667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.433990955352783, "target_z": -0.24804678832026084, "ce": 6.117925643920898, "genuineness": 4.68307638168335, "identity": 0.7405139207839966, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/1347a892808ff4331506bca9970ca595d5815f3178fda7f464bbb97dadc17d61.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "1359c9360bf31676ebe513a36c0e49312def6cd9958696fb4e90aacab127a9d1", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "257b4dba04ba6e28d6d53628c3a06f5f38f21e236bdeca79a1061905f374bec3", "generation_seconds_batch": 20.199001275002956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.288743019104004, "target_z": 0.421874849633463, "ce": 5.470598220825195, "genuineness": 4.507347106933594, "identity": 0.6432514190673828, "hypothesis": "Bon, 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_CASU_reg/1359c9360bf31676ebe513a36c0e49312def6cd9958696fb4e90aacab127a9d1.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "13919b363e87c22479b616e2c4eb548f79ec7d420b536c87009d82035cbbf56b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "f825c35db12f82c78ac10b8e775930636636bc3108a6cefdbc0aa8f411b58fd4", "generation_seconds_batch": 23.224070479045622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.184079647064209, "target_z": 0.3398437430656209, "ce": 6.181467056274414, "genuineness": 4.406930446624756, "identity": 0.45453163981437683, "hypothesis": "I found the little box. It was beside the old window. We can look inside together?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/13919b363e87c22479b616e2c4eb548f79ec7d420b536c87009d82035cbbf56b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "1405ea39876e15d73a511f52cacc97fb93ba4218f122ad0947323e21705989b5", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "b9667a57bd55251ce2bc3bb81a6a175e600319420b2c87e663f67e7b13dae393", "generation_seconds_batch": 18.85998826404102, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.017116546630859, "target_z": 0.2089845167898049, "ce": 5.967014789581299, "genuineness": 4.720732688903809, "identity": 0.707350492477417, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/1405ea39876e15d73a511f52cacc97fb93ba4218f122ad0947323e21705989b5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "1468c5f5ee37175683fb4c35ebed5371deca8f52cd94e899a4bc1e34e340f82c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "13aead6fac9f125bad80d4b8a359d3a8314d3db306daa563d266a3e7f8a0e9da", "generation_seconds_batch": 27.393089324003085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.255280017852783, "target_z": -1.1718748729913717, "ce": 6.086154937744141, "genuineness": 4.356721878051758, "identity": 0.5554256439208984, "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_CASU_reg/1468c5f5ee37175683fb4c35ebed5371deca8f52cd94e899a4bc1e34e340f82c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "149689d605ea0efe74d22a856d83d3d7d8d8593c0ebbd9751a2ef97fd26f4b67", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "8fe2f2c1dd245e1ed51de16ba6c46b17962998458cd1e6e371d8bb25ecfe306b", "generation_seconds_batch": 23.857949275989085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.333599090576172, "target_z": 0.45703130511006573, "ce": 5.831989288330078, "genuineness": 4.406930446624756, "identity": 0.6221154928207397, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/149689d605ea0efe74d22a856d83d3d7d8d8593c0ebbd9751a2ef97fd26f4b67.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "69d4d4c4c40a384bfbc9d284dc052097533e36d59d49be9bfb9eee37bcd59554", "generation_seconds_batch": 18.3169145709835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1655683517456055, "target_z": -1.2421874102180386, "ce": 6.149696350097656, "genuineness": 4.231201171875, "identity": 0.5657716393470764, "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_CASU_reg/15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "15b02c6019b1d94dfc56e6623c42f9d27d218db880fa980a215680e66b704279", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "e0caf3c01bc5d0507d9c2649683fc4d007f45d087aa9b45ce046c2457b2409ef", "generation_seconds_batch": 16.04218107496854, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.660761833190918, "target_z": -0.07031253722666692, "ce": 6.157639026641846, "genuineness": 4.432034492492676, "identity": 0.40992265939712524, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/15b02c6019b1d94dfc56e6623c42f9d27d218db880fa980a215680e66b704279.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": "16cf512040eeff239d7ad2ccb5b714ae792c006d91e6d45b230bae24149437f9", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3d22686e2ef065ac82214944b62511c05f4ae534913cc5a122f4465cf8e557e", "generation_seconds_batch": 24.002198294969276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7123148441314697, "target_z": -0.02990727938011474, "ce": 5.792276382446289, "genuineness": 4.720732688903809, "identity": 0.8103481531143188, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Le bien vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.09090909090909091, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/16cf512040eeff239d7ad2ccb5b714ae792c006d91e6d45b230bae24149437f9.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": "173e8b5e68b89ff823e12ed02a1306ee870dae546aac35b0133531e6a3c19383", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5797a7bcbe094be124fdca6adaeb521378a9a6f0445ccc5619b438d33064c217", "generation_seconds_batch": 16.813586806994863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.704371690750122, "target_z": -0.0361328160584314, "ce": 5.665193557739258, "genuineness": 4.2374773025512695, "identity": 0.6074063181877136, "hypothesis": "Ik schreef het hem naar op de envelop, lees het goed en controleer de laatste twee cijfers. Bel mij wanneer je even tijd hebt.", "wer": 0.13043478260869565, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/173e8b5e68b89ff823e12ed02a1306ee870dae546aac35b0133531e6a3c19383.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "17a98f92684b60cb9a91a16eccc1b9c868edf2a650a214fe46d9a5e171c8dbd2", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "f17d3f5dbd2fb8474a40953fdc3f5895af3cca41328b90af939765c567b45339", "generation_seconds_batch": 23.617501189000905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.296931505203247, "target_z": -0.3554686890504571, "ce": 5.689021587371826, "genuineness": 4.645420074462891, "identity": 0.7252602577209473, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/17a98f92684b60cb9a91a16eccc1b9c868edf2a650a214fe46d9a5e171c8dbd2.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "17d7aaadbebf570e92f0286aa850a5396ea067ea724e720bf9c774d4a6e8da18", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "691f497036cec9fd057f361217aac61703017dfb15ae88951e49e93bb0974396", "generation_seconds_batch": 16.82583533693105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1075401306152344, "target_z": -0.5039063299278437, "ce": 6.141753673553467, "genuineness": 4.507347106933594, "identity": 0.4625084400177002, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/17d7aaadbebf570e92f0286aa850a5396ea067ea724e720bf9c774d4a6e8da18.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "1801b99648d82a2729f7740b727989404985cfac5826cd629ad00aefc398ebab", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "88ffff933ebf4b3b660c0de80d2b4064b295f50b1ff68e7398375be575b7f3ab", "generation_seconds_batch": 18.34246405295562, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.593478202819824, "target_z": -0.12304684671503249, "ce": 6.070269584655762, "genuineness": 4.231201171875, "identity": 0.6946591734886169, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/1801b99648d82a2729f7740b727989404985cfac5826cd629ad00aefc398ebab.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": "18a38af0d519cf4d0bdad22a378a3fd9a7f7e0c2935b4b84586202f6876151b1", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The road turns left behind the bridge.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6bec8ee1a803d20036603dffcce6cdca09c7e62f192333baef3311a7070ac6a4", "generation_seconds_batch": 22.062792692915536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.106827735900879, "target_z": 0.2792966802899333, "ce": 6.300607204437256, "genuineness": 4.6203155517578125, "identity": 0.49362069368362427, "hypothesis": "The road turns left behind the bridge. Keep going. Until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/18a38af0d519cf4d0bdad22a378a3fd9a7f7e0c2935b4b84586202f6876151b1.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": "18b5a28d06ca37808767fd4d7f0519d2a8522a9c71a1520642e8f5f08dd14a3d", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "362a41a1261d289baabb278c3e689e25b4e23a43f89251409809e268d915da24", "generation_seconds_batch": 24.24899140105117, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.622136116027832, "target_z": -0.10058588175124145, "ce": 6.046441555023193, "genuineness": 4.469690799713135, "identity": 0.7127470970153809, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de daarna naast het raam, neem ze mee als je weggaat.", "wer": 0.09090909090909091, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/18b5a28d06ca37808767fd4d7f0519d2a8522a9c71a1520642e8f5f08dd14a3d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "18d495b374f4c60a00ae4e834b6ea6f3646f3267fcf06f43db9dc36e52c0ab14", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "be55aaca493882a95dd9e8201e8659069e923787188701d507341c03cf2fec58", "generation_seconds_batch": 24.70952018199023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0527164936065674, "target_z": -0.5468749781019606, "ce": 5.927301406860352, "genuineness": 4.7332844734191895, "identity": 0.6899884939193726, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/18d495b374f4c60a00ae4e834b6ea6f3646f3267fcf06f43db9dc36e52c0ab14.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": "195f7d03b8577db179bb770dcadc0da54bad1311d28822d0cba0800b6f9f8806", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "12c8316a974a9ecfc235f75378a18a74036d75a761481271c032602679c15e00", "generation_seconds_batch": 23.178023271961138, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.914944887161255, "target_z": 0.12890631824888935, "ce": 5.967014789581299, "genuineness": 4.444586753845215, "identity": 0.5992470383644104, "hypothesis": "Las llaves están en la mesa de la cocina. Miren el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.08695652173913043, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/195f7d03b8577db179bb770dcadc0da54bad1311d28822d0cba0800b6f9f8806.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "19770fd4d6b8b2d38060cc992f7d9887c3f433f82f62cc4647038494cd6ba2bc", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "7c6facc8722b859ce2c5745983f02306e7450fdc7101ff754487dc9bf478c634", "generation_seconds_batch": 22.726629375945777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.863325357437134, "target_z": -0.695312432116078, "ce": 6.070269584655762, "genuineness": 4.557555198669434, "identity": 0.6547629833221436, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/19770fd4d6b8b2d38060cc992f7d9887c3f433f82f62cc4647038494cd6ba2bc.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "1998c882d2843d0a823f845d9b6c354d42b28c94814fc8dc61c5c6ecf8aaf5a3", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "cd94658cbfe23ced8f80945d8afb25ddd7e4b3ee0fbffd5a738a61eb6e6be31c", "generation_seconds_batch": 16.66787780507002, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9562411308288574, "target_z": -1.4062499970802613, "ce": 6.006728172302246, "genuineness": 4.268857479095459, "identity": 0.7393471002578735, "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_CASU_reg/1998c882d2843d0a823f845d9b6c354d42b28c94814fc8dc61c5c6ecf8aaf5a3.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": "19bed21de4c317f0285101513c00d90e786e32de2296039f0f7598967a0e4a49", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The road turns left behind the bridge.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eed384be9147c0bb54fb797e43041bfdafe31f1a2cdd2ac9c3a73feebc573e01", "generation_seconds_batch": 17.88598236395046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0577003955841064, "target_z": -0.5429687883215689, "ce": 6.213237762451172, "genuineness": 4.193545341491699, "identity": 0.5418664216995239, "hypothesis": "The road turns left behind the bridge. Um keep going until you reach the 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_CASU_reg/19bed21de4c317f0285101513c00d90e786e32de2296039f0f7598967a0e4a49.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "1a2fa7dba5f08983e97cec8df78774dfe51b1a22110da4d5ba2796e1a5cb445f", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "df022ca76cac470318cbca8a8b74ce4d7d9a2b2fc621e5fa606b38242241baaa", "generation_seconds_batch": 22.555652193026617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.677493572235107, "target_z": 0.7265625109490197, "ce": 5.824047088623047, "genuineness": 4.570107460021973, "identity": 0.6142618656158447, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/1a2fa7dba5f08983e97cec8df78774dfe51b1a22110da4d5ba2796e1a5cb445f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "1a5cc59329d17ca4ef1fa9d98cd351d2e752367c160b26439903dbe164bcb6e2", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "8587fb1fa380ed99939bfe5577a92b795c5279b3f4279a4ac6b9fbabdbcc65f6", "generation_seconds_batch": 17.752890571951866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.566066265106201, "target_z": -0.1445312642337256, "ce": 6.38003396987915, "genuineness": 4.6705241203308105, "identity": 0.3731434643268585, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/1a5cc59329d17ca4ef1fa9d98cd351d2e752367c160b26439903dbe164bcb6e2.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "1aebdff2b38a1ccddf9ddec36718102896bce8619668c7c082d50bd1c162e007", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3f51e2f5cbc1cc8b3bf348df5ffb4470fd94a5064eba35e5b6dab5485bbd3243", "generation_seconds_batch": 22.053056915057823, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.149191856384277, "target_z": 0.31250004087634015, "ce": 5.959072113037109, "genuineness": 4.607763767242432, "identity": 0.4525797665119171, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/1aebdff2b38a1ccddf9ddec36718102896bce8619668c7c082d50bd1c162e007.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": "1b3c397e956630879e2dcb577207a7aeae72c92c9069d2ca100086b2621e65f8", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5491d0437c47973e68c59c6259ae312fb11ad684e012324154d2e2812c19af85", "generation_seconds_batch": 22.40629797708243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.90559983253479, "target_z": 0.12158202554738558, "ce": 5.80021858215332, "genuineness": 4.595211505889893, "identity": 0.643565833568573, "hypothesis": "Les clés sont sur la table de la cuisine. J'y regardai dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras?", "wer": 0.08333333333333333, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/1b3c397e956630879e2dcb577207a7aeae72c92c9069d2ca100086b2621e65f8.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "1bf2d72809e16092d532c951691c2aea7abad613cbffa2c849ce44b178762120", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "243ec01227e8a2d8517438aefd6af22ec5c6033556d88029203b9d7f592c37e2", "generation_seconds_batch": 24.704312894959003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4745750427246094, "target_z": -0.9999999065683653, "ce": 5.903473377227783, "genuineness": 4.337893962860107, "identity": 0.6638196706771851, "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_CASU_reg/1bf2d72809e16092d532c951691c2aea7abad613cbffa2c849ce44b178762120.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": "1c00c00c79ac320780fe7731726822ebde5e956955aa4bfba7642c1f2146ef4c", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa685096fe21f213d948104b0fd27d87264c61bba915b0c1cfc7fdabaefffad7", "generation_seconds_batch": 15.38395607098937, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.737301349639893, "target_z": 0.7734375357667976, "ce": 5.911416053771973, "genuineness": 4.557555198669434, "identity": 0.6262060403823853, "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_CASU_reg/1c00c00c79ac320780fe7731726822ebde5e956955aa4bfba7642c1f2146ef4c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "1cfb19d5c6079971315ecf6f3a34f197516b8e8fd25897e8e9881ee17343abac", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "05b513c501760c486c84108797335fb76919bbf882ec5a68ba9f9c6fab574559", "generation_seconds_batch": 26.84206254093442, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.697429656982422, "target_z": 0.7421876437971252, "ce": 6.229123115539551, "genuineness": 4.632867813110352, "identity": 0.31180962920188904, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/1cfb19d5c6079971315ecf6f3a34f197516b8e8fd25897e8e9881ee17343abac.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": "1d04b3359865e73d9d0dc427f274adea116545c54ca43d648035a2cf37ccad87", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff8bcbea722373728c5f4b96a16a214758adca66fd54c1af239ed958f5965ec5", "generation_seconds_batch": 17.036728291073814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9530367851257324, "target_z": -0.6250000817526803, "ce": 5.903473377227783, "genuineness": 4.394378662109375, "identity": 0.8186477422714233, "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_CASU_reg/1d04b3359865e73d9d0dc427f274adea116545c54ca43d648035a2cf37ccad87.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": "1d224bdb73ac8049c0278388f969d64f66e58a92f32fca985c30fb1f71227d77", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bd5dbe3804a563216fb176ed249783a7c33b757376dd03494ffee22721836137", "generation_seconds_batch": 17.656199467019178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6345958709716797, "target_z": -0.09082040730026208, "ce": 6.078212261199951, "genuineness": 4.419482707977295, "identity": 0.5656745433807373, "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_CASU_reg/1d224bdb73ac8049c0278388f969d64f66e58a92f32fca985c30fb1f71227d77.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "c531069f6b2d913d122ce23fcbf9e2848967cccbf9aa8c1b4b68b942bcfc0b02", "generation_seconds_batch": 17.90887283894699, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.185504198074341, "target_z": -1.2265624642332025, "ce": 6.221180438995361, "genuineness": 4.13078498840332, "identity": 0.7613908052444458, "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_CASU_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": "1d5f76882ca2a502b52ba41c8cf9c1aa2ee4d2fab0dc3f64b0bfdcbfb83ee389", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "12d79d94a5f7fadd0427261b1818fbe7abcea19953686ccf2327018865b6450d", "generation_seconds_batch": 22.545689053018577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0028767585754395, "target_z": -0.5859374364956859, "ce": 6.062326908111572, "genuineness": 4.469690799713135, "identity": 0.7174102067947388, "hypothesis": "Drei Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "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_CASU_reg/1d5f76882ca2a502b52ba41c8cf9c1aa2ee4d2fab0dc3f64b0bfdcbfb83ee389.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "1d7acf99a4ccfab0710607724305c03b1b5c03c62fb3acfc22dcfea2270e6812", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "3a02f01c8cf02c758486b6aa1d598d777f68be381e64be12690df4399822b598", "generation_seconds_batch": 26.635256957029924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4713706970214844, "target_z": -0.21874999124078426, "ce": 6.09409761428833, "genuineness": 4.6203155517578125, "identity": 0.7318424582481384, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/1d7acf99a4ccfab0710607724305c03b1b5c03c62fb3acfc22dcfea2270e6812.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "f4b336bbe80929183df54eb5a54421d8827880e6884de15212c588a626843251", "generation_seconds_batch": 16.863939173985273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.658982276916504, "target_z": -0.8554686423346398, "ce": 6.332377910614014, "genuineness": 4.293961524963379, "identity": 0.5994157791137695, "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_CASU_reg/1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "1dcd1df8c632c888492460e1d2b0564c81ec8d56716d598be9a6b0d5a54e5ac8", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "8d1d7638d8c3d01ea34fd175dd1b0bf0068322dc24c905c2b3b28c438fd999bc", "generation_seconds_batch": 26.80855601897929, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8280367851257324, "target_z": 0.06079110620532741, "ce": 6.300607204437256, "genuineness": 4.337893962860107, "identity": 0.3259119987487793, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/1dcd1df8c632c888492460e1d2b0564c81ec8d56716d598be9a6b0d5a54e5ac8.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "1e2389401e8133728ad262ff110e08b492a0f2ea992e029e92ca29d1b2364042", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4d80bbc6409778db8cde21780cc0d855952e1a94ed2f0d1c86deee32e18cf58c", "generation_seconds_batch": 17.461880658054724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6956496238708496, "target_z": -0.0429688350373862, "ce": 6.165581703186035, "genuineness": 4.369274139404297, "identity": 0.49041515588760376, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/1e2389401e8133728ad262ff110e08b492a0f2ea992e029e92ca29d1b2364042.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "f4a02f43ed7cb88ffe03cd700e68d1aa3d671b2e524af66afb06ed617a6cd43b", "generation_seconds_batch": 17.90887283894699, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1847920417785645, "target_z": -0.4433592671521561, "ce": 6.2767791748046875, "genuineness": 4.394378662109375, "identity": 0.720391035079956, "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_CASU_reg/1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8.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": "1fb22ebbe2a58f7847a89dc54df5a4fc5b0b790ac8d14d224454f0fd9358cf47", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4170869f74c1d8d46e803a54f82433153c540c4f39cb796ccc3a49d9f06a9f39", "generation_seconds_batch": 25.32326394005213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.966564178466797, "target_z": 0.9531248817505874, "ce": 5.927301406860352, "genuineness": 4.469690799713135, "identity": 0.19572526216506958, "hypothesis": "De weg gaat linksaf naar de brug. Loopt verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/1fb22ebbe2a58f7847a89dc54df5a4fc5b0b790ac8d14d224454f0fd9358cf47.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "1fdcffc6fb42aaadd7ff77ecd49bade1bac569e498510a3a205f7f004f4a6bd0", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe2f71ee081e453730df5ba057c410aa601ad8fb0c32a70a6bae2b5facef7586", "generation_seconds_batch": 23.322285901056603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.617685794830322, "target_z": 0.6796874861312417, "ce": 6.070269584655762, "genuineness": 4.243753433227539, "identity": 0.7831976413726807, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/1fdcffc6fb42aaadd7ff77ecd49bade1bac569e498510a3a205f7f004f4a6bd0.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "2001fa55d44c7510b88ca904b4744599a8e81a4a30d5142968e6501935486c40", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 128, "seconds_generated": 10.24, "mp3_sha256": "ee5d51e45e9c0206656f4665168a2110fb12844d2ddee6cf16678cd9cf951ae1", "generation_seconds_batch": 21.90097497892566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.624094247817993, "target_z": -0.8828125313871897, "ce": 6.308549880981445, "genuineness": 4.432034492492676, "identity": 0.5110100507736206, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/2001fa55d44c7510b88ca904b4744599a8e81a4a30d5142968e6501935486c40.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "200b846810026038d070de2ab6afb21656c329398c78e9bde99f2a615ab5f51f", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "1556e1ee73bc7f55500adaae1d5c1ad613ee623a094a8129efd4dc1b8922ff73", "generation_seconds_batch": 17.171801276039332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.032780647277832, "target_z": -0.5624999240867968, "ce": 6.070269584655762, "genuineness": 4.293961524963379, "identity": 0.5836324691772461, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/200b846810026038d070de2ab6afb21656c329398c78e9bde99f2a615ab5f51f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "20f43d18f0bac7336ce44204a699345701b7732b5e327e3aae39c2d205884b6e", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "2a84591e39ac6ceb12f85d16aaf1c814ae0cc9cf0f06c98a4c93d6bd8772e773", "generation_seconds_batch": 16.297755577019416, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8166897296905518, "target_z": -1.5156249927006535, "ce": 6.109982967376709, "genuineness": 4.381826400756836, "identity": 0.3958793878555298, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/20f43d18f0bac7336ce44204a699345701b7732b5e327e3aae39c2d205884b6e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "213f3831a2607059cd24be0c08c7543f4ca1e2642e0c79e7522ce4ad3a9952c1", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "23b1d22b3c2ff37fdd96bec497407b06ec342c1aa96979ba01289f1899e9c553", "generation_seconds_batch": 23.965668428922072, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6084301471710205, "target_z": -0.111328090510588, "ce": 5.935244083404541, "genuineness": 4.632867813110352, "identity": 0.6854413747787476, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/213f3831a2607059cd24be0c08c7543f4ca1e2642e0c79e7522ce4ad3a9952c1.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "21832da97537aef6c0f202819b39a5b0858345488e57a4ec72cb3a67aeed435b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "15eddc58e9720b3d5c6e5f9d7e2b354ac36783428c511d17e722a13511a63749", "generation_seconds_batch": 27.393089324003085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5523602962493896, "target_z": -0.15527347299307215, "ce": 6.0305562019348145, "genuineness": 4.432034492492676, "identity": 0.6337196230888367, "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_CASU_reg/21832da97537aef6c0f202819b39a5b0858345488e57a4ec72cb3a67aeed435b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "21f577f4a660a1279456f40555416e351e07b47bae6709363f7d6d6396d110b4", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "00d0ca494706b280f6134de1822ce3cb0a0e796693973a1a2789c555ca1c72c4", "generation_seconds_batch": 25.41339203598909, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.36919903755188, "target_z": -0.2988281897816998, "ce": 5.7684478759765625, "genuineness": 4.6203155517578125, "identity": 0.7929183840751648, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/21f577f4a660a1279456f40555416e351e07b47bae6709363f7d6d6396d110b4.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "223bc23286412ec253c3a3b7d001bc9b9e08b203b14f739a72cc58f18e5285d4", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f240be83096b7d6e6e831e8942dad7314a63fd907d46dec55a75ebd1b3669616", "generation_seconds_batch": 25.096467626048252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0975723266601562, "target_z": -0.5117187094886272, "ce": 6.24500846862793, "genuineness": 4.143336772918701, "identity": 0.3248039484024048, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/223bc23286412ec253c3a3b7d001bc9b9e08b203b14f739a72cc58f18e5285d4.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "2260942bd588991323deedadc65500589d26b187b4c598e6514d74f2aadbc885", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "fa46eae3fee726eb28233cfcf9f9008750722a65dbb1a1c4e6efcb202b077c44", "generation_seconds_batch": 23.066818468039855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1523959636688232, "target_z": -0.46875006131451025, "ce": 6.09409761428833, "genuineness": 4.545003414154053, "identity": 0.41953665018081665, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/2260942bd588991323deedadc65500589d26b187b4c598e6514d74f2aadbc885.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "22df1ec8788459519c212e71d62db70ff4d4df74e738a5757b4052164c90e85a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "abe341e7c22f3edce56c816d87bfcab337f05fe1a47d9e9a60dee132dde5411e", "generation_seconds_batch": 18.040603850968182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.104335784912109, "target_z": 0.2773435853997374, "ce": 6.0146708488464355, "genuineness": 4.432034492492676, "identity": 0.7730504274368286, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/22df1ec8788459519c212e71d62db70ff4d4df74e738a5757b4052164c90e85a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "22f0e72056cfefeeee1de1a2d1e385caceabc1a7e81354bc8d90b97d68a78186", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "55b832b90b4a4ce280ef17f6d440b3d1ddb88b0618d33528d2761bc83fe187ab", "generation_seconds_batch": 17.177611363935284, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.196539402008057, "target_z": 0.34960921751660023, "ce": 6.252951145172119, "genuineness": 4.6956281661987305, "identity": 0.3566880226135254, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/22f0e72056cfefeeee1de1a2d1e385caceabc1a7e81354bc8d90b97d68a78186.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "22fded2db9e89c0df55cd42a725910db0e9333a4decda7da3436c8c47fca8003", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e756617c6c35adbd6f82b6776ba440db7799f9da5ad7ce57b9ae5c0711d3f015", "generation_seconds_batch": 22.547551158932038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.171619892120361, "target_z": 0.3300782686146415, "ce": 6.165581703186035, "genuineness": 4.507347106933594, "identity": 0.43363213539123535, "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_CASU_reg/22fded2db9e89c0df55cd42a725910db0e9333a4decda7da3436c8c47fca8003.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "2309951cb431ee540961814283579b8c434eb2f0ceb6802ce4715f1d61182a50", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "22dab3fc3898e99e0d0a9d3df5df0dfb816929db3057e4c51a9aeb18d9b885c6", "generation_seconds_batch": 22.328026499948464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9897046089172363, "target_z": 0.18750009927111178, "ce": 5.847874641418457, "genuineness": 4.369274139404297, "identity": 0.517645001411438, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/2309951cb431ee540961814283579b8c434eb2f0ceb6802ce4715f1d61182a50.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "23af0fc0f49f7bebbec5c857707beb59452556221bf2fafd1a66bd69fbbcf7ef", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e4ee937ec8b83c6089e72291292586d8027e4848f6a9e19b73301e9bdd3e8a8c", "generation_seconds_batch": 22.56079834396951, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0769243240356445, "target_z": 0.25585954160758284, "ce": 5.911416053771973, "genuineness": 4.444586753845215, "identity": 0.39372915029525757, "hypothesis": "I found the little box it was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/23af0fc0f49f7bebbec5c857707beb59452556221bf2fafd1a66bd69fbbcf7ef.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "25509f2b9a870ebbaf3dc558652db5a2a2538cff392dc74aaadb4901bd4c876c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c0957879e930be96db2fe8e47611f88b612d79e0b3aba41f06ad3736e72865dd", "generation_seconds_batch": 21.95207603799645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0975723266601562, "target_z": -0.5117187094886272, "ce": 5.919358730316162, "genuineness": 4.6705241203308105, "identity": 0.7102483510971069, "hypothesis": "J'ai trouvé la petite boîte et 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_CASU_reg/25509f2b9a870ebbaf3dc558652db5a2a2538cff392dc74aaadb4901bd4c876c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "25b5101bfb3e2d2a17f1cd6ff58edd1ca67dde33c5ec071f83b7843b3ad789d8", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "6bb98afe4fd83be3b218626d3c9e44d89ae836e3792959b24d7e7616cae170a1", "generation_seconds_batch": 17.30407881201245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.992196559906006, "target_z": 0.18945319416130765, "ce": 5.7684478759765625, "genuineness": 4.256305694580078, "identity": 0.4669520854949951, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/25b5101bfb3e2d2a17f1cd6ff58edd1ca67dde33c5ec071f83b7843b3ad789d8.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "260b0dff2ecd2ae070ee56e91f502bd5b2a4cc77aca8336f878f623269cf7c05", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6ae529a3f2305f31deab1915eae1866a0bfb1e4c217476fae5d4cab912e5c00c", "generation_seconds_batch": 16.244926756015047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.293727397918701, "target_z": 0.42578141314039325, "ce": 5.474569320678711, "genuineness": 4.6705241203308105, "identity": 0.6085920929908752, "hypothesis": "J'ai trouvé la 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_CASU_reg/260b0dff2ecd2ae070ee56e91f502bd5b2a4cc77aca8336f878f623269cf7c05.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "262a4688018d00afa361fa9fa9924c5c785a24f960f41c77008ec831333e820e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "a781bdcc07bd1e2118378021153432e5147c2af4c533efe1f700f06946cf0c88", "generation_seconds_batch": 22.463291352032684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.348550796508789, "target_z": 0.468749874451241, "ce": 5.998785495758057, "genuineness": 4.519899368286133, "identity": 0.36751028895378113, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/262a4688018d00afa361fa9fa9924c5c785a24f960f41c77008ec831333e820e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "263adcdf7282e7dc2851a08767653e8a8a2d2e6ab217be7064252efbac1ab725", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "d97472b61374e6f7933edbdb86d16361d8b74e2bac689e823b5186dfeb02430f", "generation_seconds_batch": 21.66386050102301, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.134239673614502, "target_z": 0.3007810978086264, "ce": 5.982900142669678, "genuineness": 4.720732688903809, "identity": 0.8337207436561584, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. Pudemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/263adcdf7282e7dc2851a08767653e8a8a2d2e6ab217be7064252efbac1ab725.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": "268023a0dd1f2356a827d9a5a7eb58677b900ba4f4255495bcd7a66e360216e8", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "62d6b205b6018607f6694494a1dedf30b997d99fdd2cb03dcce9444d4de65749", "generation_seconds_batch": 16.813586806994863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.675090789794922, "target_z": -0.0590820547447714, "ce": 6.006728172302246, "genuineness": 4.457139015197754, "identity": 0.42186057567596436, "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_CASU_reg/268023a0dd1f2356a827d9a5a7eb58677b900ba4f4255495bcd7a66e360216e8.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "2681b1b365b8a1eb2b6a9a0f8b54ca92fc5af71bc4fdf0ef2c71a4457064213a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "9e97c5ed19a912a205ab550690ca787c1b24649d7ec363675355f9e5c00041ec", "generation_seconds_batch": 22.38563629903365, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.792125225067139, "target_z": 0.8164063708041839, "ce": 5.998785495758057, "genuineness": 4.632867813110352, "identity": 0.21708782017230988, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/2681b1b365b8a1eb2b6a9a0f8b54ca92fc5af71bc4fdf0ef2c71a4457064213a.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": "26aed215c227a67d5c7f409b70c48f89fd7ffac2c609ecc82a6446d0c83eca95", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a87ccaa728a35eef11fdf486ce779ff755c53bf14d59aa53d69a27ab514d5c77", "generation_seconds_batch": 22.26796659291722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4439587593078613, "target_z": -0.24023440875947735, "ce": 6.070269584655762, "genuineness": 4.6203155517578125, "identity": 0.5651307106018066, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann leg dich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.08333333333333333, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/26aed215c227a67d5c7f409b70c48f89fd7ffac2c609ecc82a6446d0c83eca95.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "26b757aac75cbbcbeff2792d28f71f6bf2f6578b2d2e25387784a142defd4a21", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "4554974a828a227d26480c3fd577289f665ed1d16710cbeca4c0313b111796ba", "generation_seconds_batch": 25.300991802942008, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.149191856384277, "target_z": 0.31250004087634015, "ce": 5.903473377227783, "genuineness": 4.413206577301025, "identity": 0.6471834182739258, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/26b757aac75cbbcbeff2792d28f71f6bf2f6578b2d2e25387784a142defd4a21.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 127, "seconds_generated": 10.16, "mp3_sha256": "eae6023a052120c26c2abb9e5b5c0425c0d949656d5714bf7209636d23f4d3a8", "generation_seconds_batch": 17.3396077939542, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8035175800323486, "target_z": -0.7421874569338559, "ce": 6.3641486167907715, "genuineness": 4.419482707977295, "identity": 0.16986426711082458, "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_CASU_reg/26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "27146f87b6312cf52aa848fba9a3fee27481f86d5625b10774d5f537a653657a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "46b8b47806ace1e2a428dcf7d58905d69af825faa2683a3573f01c12437ccce1", "generation_seconds_batch": 21.620909058023244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.823453426361084, "target_z": -0.7265625109490197, "ce": 6.141753673553467, "genuineness": 4.15588903427124, "identity": 0.2892875671386719, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/27146f87b6312cf52aa848fba9a3fee27481f86d5625b10774d5f537a653657a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "27ca9fb985a2cad3bea532c3ce56b492bc7877983e386779838fd217e712dbfd", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "055159b869be261bad2690320c3ec2e35f341446a9ec26d8490be9ee0094a230", "generation_seconds_batch": 24.32315170194488, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.856561541557312, "target_z": -1.4843750072993465, "ce": 6.348263263702393, "genuineness": 4.507347106933594, "identity": 0.49895745515823364, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/27ca9fb985a2cad3bea532c3ce56b492bc7877983e386779838fd217e712dbfd.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "280e73046d080be74e6b6d91149e05e006a84f049b18470cc2fce2cca34c4049", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a1cda1616d15031c05ab18c78cc7e573c8f58f50cd44c918f6f930a4bdaf841", "generation_seconds_batch": 22.471012047026306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4265148639678955, "target_z": -0.2539062598541177, "ce": 6.109982967376709, "genuineness": 4.758388519287109, "identity": 0.729863703250885, "hypothesis": "La porte est encore ouverte j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/280e73046d080be74e6b6d91149e05e006a84f049b18470cc2fce2cca34c4049.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "280e7e64a8560c360dd706a5a005071a5a8b533b4af1cd7d3b7d4746d4ae2f10", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "23a30ccae324c83ec69bc57d8790f08a1117e5070d85687961bf96b66027cab2", "generation_seconds_batch": 22.763790930970572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.717365264892578, "target_z": 0.7578124029186921, "ce": 6.157639026641846, "genuineness": 4.65797233581543, "identity": 0.2924101948738098, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/280e7e64a8560c360dd706a5a005071a5a8b533b4af1cd7d3b7d4746d4ae2f10.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "283a426697e97e2a6542c6f4a09add87cb343e74466602202662d58d867a3b6f", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "7d07c7cc1693c7c23230d586931885063f71751dc0a009d834c4889280ff47b5", "generation_seconds_batch": 21.246189445024356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.085824489593506, "target_z": -1.304687567883922, "ce": 6.41180419921875, "genuineness": 4.231201171875, "identity": 0.6051743030548096, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/283a426697e97e2a6542c6f4a09add87cb343e74466602202662d58d867a3b6f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "288e4b3c3a16e55913a81c2dab011ebcfa4b078c76e03e6155486d77d08ce455", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "bebc7c5f96ac1833c5eea6a2d822dd531c5e602610c48cd6763b30f753330402", "generation_seconds_batch": 22.007058573886752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1798079013824463, "target_z": -0.4472656437958171, "ce": 5.7684478759765625, "genuineness": 4.632867813110352, "identity": 0.6723837852478027, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/288e4b3c3a16e55913a81c2dab011ebcfa4b078c76e03e6155486d77d08ce455.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "292b5e2037549c37a53a5b8e9ff8c47dfd81e53ef34d5679fe35b5bc7c756352", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "12131bc15f0d123b44fbb94d77555e9e6ee92bf8115486c8171c804c203bafb6", "generation_seconds_batch": 16.061606397968717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4813387393951416, "target_z": -0.2109374248167315, "ce": 5.4666266441345215, "genuineness": 4.645420074462891, "identity": 0.6749753952026367, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/292b5e2037549c37a53a5b8e9ff8c47dfd81e53ef34d5679fe35b5bc7c756352.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "2949371477ef6f9b9d58b105649f8e7b3e7743cc3e3c159ca9edc357361ecf83", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eccef2452ecda85537aa7dc97fe9aa783551f3f05f13b9d84215defc7d46f967", "generation_seconds_batch": 17.14279102499131, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9031970500946045, "target_z": -0.6640625401464055, "ce": 6.062326908111572, "genuineness": 4.457139015197754, "identity": 0.6741446256637573, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/2949371477ef6f9b9d58b105649f8e7b3e7743cc3e3c159ca9edc357361ecf83.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "298e036ccb3469255d7f0ab9d7cb965903d720b2ae55676fe69b75480058a91c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "277f3b4dc4b6314f4a8981bd82f6dce6812a9d6073bbf801685219f3b813427e", "generation_seconds_batch": 23.138383396086283, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.233919620513916, "target_z": 0.3789063883226153, "ce": 5.871703147888184, "genuineness": 4.6203155517578125, "identity": 0.4944668412208557, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/298e036ccb3469255d7f0ab9d7cb965903d720b2ae55676fe69b75480058a91c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "2993332b5ea80f79445692dec4c6da56217cc46754ef20530842e83923e41b94", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "59f8f1a73a227e19a66e1f72d7068df740c0b23fff29f52fd9c9f67945163f94", "generation_seconds_batch": 23.130981167894788, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.92313289642334, "target_z": -0.6484375941615693, "ce": 6.117925643920898, "genuineness": 4.300237655639648, "identity": 0.35076379776000977, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/2993332b5ea80f79445692dec4c6da56217cc46754ef20530842e83923e41b94.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "2a390465bba8d269ed80a33ac7bfb68dfefd7d1f4497b0c2ea093868f7882d80", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "6f427cdd6fbb8cfe2874b8aa9231adfdbd52437342d19bb793abafadab49a9ca", "generation_seconds_batch": 23.066818468039855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6420719623565674, "target_z": -0.08496093576640522, "ce": 6.09409761428833, "genuineness": 4.770940780639648, "identity": 0.6330937147140503, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/2a390465bba8d269ed80a33ac7bfb68dfefd7d1f4497b0c2ea093868f7882d80.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": "2a60803c9ccfdceb2ac242c2d86875b7c285b8b86536830467c87da30dc42cc3", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "472362faa2b2650bbae85eaf75c83e902c2083d1bb25aabf08e60e6858242da5", "generation_seconds_batch": 20.494089017040096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.503054141998291, "target_z": 0.5898436262760776, "ce": 5.879645347595215, "genuineness": 4.331617832183838, "identity": 0.12056644260883331, "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_CASU_reg/2a60803c9ccfdceb2ac242c2d86875b7c285b8b86536830467c87da30dc42cc3.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "2ab165096ebe60bcb0454f3f42ca8e3c5bad99491653fd68a2a5d24babb1ee94", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "69f99252be4a1f38d80c1ea28d9a47ee148e4ba98105800ea41bbef06f03a704", "generation_seconds_batch": 18.261779900989495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.919928789138794, "target_z": 0.1328125080292811, "ce": 6.070269584655762, "genuineness": 4.419482707977295, "identity": 0.14224420487880707, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/2ab165096ebe60bcb0454f3f42ca8e3c5bad99491653fd68a2a5d24babb1ee94.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "2ab9083207397078f1cb7cf066f47723238dbbcd9120f046fdcb31b38f739d45", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "ea64c146fbb9be2b86522c828deac0b1d91073649de3cdb9ce7ce7393f874fae", "generation_seconds_batch": 23.764447032939643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.306899309158325, "target_z": -0.34765630948967363, "ce": 5.927301406860352, "genuineness": 4.607763767242432, "identity": 0.476326584815979, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/2ab9083207397078f1cb7cf066f47723238dbbcd9120f046fdcb31b38f739d45.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "2b45af400a9efd7244e7723456287ea68e497fcbb13ad4015f170edab155a7b9", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e64cdd4701eac17b3a1cf04e51c717f736c52cefa119c0ecb84d4e0364308c3b", "generation_seconds_batch": 24.264147295965813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1349520683288574, "target_z": -0.48242191240915056, "ce": 5.895530700683594, "genuineness": 4.545003414154053, "identity": 0.7060103416442871, "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_CASU_reg/2b45af400a9efd7244e7723456287ea68e497fcbb13ad4015f170edab155a7b9.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "2c129ef53307a27ede6067853b59e3fb98e18c443cfd43fad93a30331813a4f7", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "9e70e60a0a4694aa96ca718ae8eb79fa5963d639e874c247ec3f05afd11c1645", "generation_seconds_batch": 21.121508384938352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1723318099975586, "target_z": -0.453125115329674, "ce": 6.046441555023193, "genuineness": 4.143336772918701, "identity": 0.35610729455947876, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/2c129ef53307a27ede6067853b59e3fb98e18c443cfd43fad93a30331813a4f7.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "2cb2f9dd0283ba41748bf94cea4aac7466efe285881b5ed8806c0bd4c09a6b4c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "edcf4789c18cb0f169edf54ffaba094779f1e590747d2903542dbb21c2a6084d", "generation_seconds_batch": 16.433814181014895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9263372421264648, "target_z": -1.4296875094891504, "ce": 6.332377910614014, "genuineness": 4.143336772918701, "identity": 0.42199650406837463, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/2cb2f9dd0283ba41748bf94cea4aac7466efe285881b5ed8806c0bd4c09a6b4c.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": "2d8148a218455ec2d8736c6c624e5164ae77245385beeb6e23ab51d9ab29428f", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c729e534ce8da68167a13eda35031c4f10f4f69e2a30cc133afbff62f0672dae", "generation_seconds_batch": 16.881773566943593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.468166351318359, "target_z": 0.5624999240867968, "ce": 5.804190158843994, "genuineness": 4.256305694580078, "identity": 0.7349539995193481, "hypothesis": "Al iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het naar de lunch terug.", "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_CASU_reg/2d8148a218455ec2d8736c6c624e5164ae77245385beeb6e23ab51d9ab29428f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "2dbcf9db2b19a7b1ce32ef2084f0c1874684b2724af53bc742a0b20bc4fa8f41", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "d76f0263d90f08a4beff766882fc0ab8360ec064bb58a95a67c531168fbad84d", "generation_seconds_batch": 25.922974191023968, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.296931505203247, "target_z": -0.3554686890504571, "ce": 6.125868320465088, "genuineness": 4.632867813110352, "identity": 0.6111409068107605, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/2dbcf9db2b19a7b1ce32ef2084f0c1874684b2724af53bc742a0b20bc4fa8f41.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "2dda5ebce2da87e594e6c7f0180b03f0c69baf595a8c15e3c8c098ebeaf2aaf6", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aa853b16940ed08a839f6052307730cfef67269ef4fb9fb4e465116e5fb5fea6", "generation_seconds_batch": 22.99366586294491, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.321139335632324, "target_z": 0.44726583065908637, "ce": 5.816103935241699, "genuineness": 4.362998008728027, "identity": 0.441213458776474, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/2dda5ebce2da87e594e6c7f0180b03f0c69baf595a8c15e3c8c098ebeaf2aaf6.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "2e0709c277ffe200456e5d5ddfc92cba3da66861855bc8c8e8309bc8d5d7b1fb", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "8604032bcecab6db7cf18a292a04656c0b1c824222a531b5c36cda5d06196398", "generation_seconds_batch": 23.1150081200758, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9274046421051025, "target_z": 0.1386717926998687, "ce": 5.879645347595215, "genuineness": 4.306513786315918, "identity": 0.31100544333457947, "hypothesis": "De deur staat nog open? Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/2e0709c277ffe200456e5d5ddfc92cba3da66861855bc8c8e8309bc8d5d7b1fb.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "2e566efb8ba1658a2157ae6c5745f82d101a721a446b916c65eabe550ef96684", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "d3108c7e406f77f0fa407d598ddd7451877c469475ca7278f3a26a0f391ea26a", "generation_seconds_batch": 25.724504982936196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.171619892120361, "target_z": 0.3300782686146415, "ce": 5.5539960861206055, "genuineness": 4.6705241203308105, "identity": 0.7297958731651306, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Eh dime cuándo estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/2e566efb8ba1658a2157ae6c5745f82d101a721a446b916c65eabe550ef96684.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "2f32cdc3054bce89e145ff0e7a0f633190e1e8f330e8fe8776d7847b84e85f1d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "83be1db44496e079104c25127dc0fa3f719f95b41f093ddb1fdc3921be2396fd", "generation_seconds_batch": 22.718367752968334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.657557487487793, "target_z": 0.7109373781009142, "ce": 5.903473377227783, "genuineness": 4.645420074462891, "identity": 0.687254786491394, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Euh dis-moi quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/2f32cdc3054bce89e145ff0e7a0f633190e1e8f330e8fe8776d7847b84e85f1d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "2f595e016dae25d474c237f1a177dcd1798fe84a978e86f1bd1bccaf1c8c4c9c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "b5a8f0f03f48363f16020a3b37b59d3d30ef821e58b3ac2bb27bc53d4c124eed", "generation_seconds_batch": 26.635256957029924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.837693214416504, "target_z": 0.06835944233647105, "ce": 6.109982967376709, "genuineness": 4.68307638168335, "identity": 0.6648904085159302, "hypothesis": "He encontrado la cajita. Estaba junto a la cre, estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.29411764705882354, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/2f595e016dae25d474c237f1a177dcd1798fe84a978e86f1bd1bccaf1c8c4c9c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "2fead0abe4d82944b4bb48f30b4ecd2fb1626181cb60d28081fc9d332b4ebefa", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "fd10b479420f371b3a0e679a13aaffa013909ef90e35a77dd1ad15cfdecbb8f5", "generation_seconds_batch": 22.368857298977673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4476966857910156, "target_z": -0.23730476642418355, "ce": 6.141753673553467, "genuineness": 4.8588056564331055, "identity": 0.6143848896026611, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/2fead0abe4d82944b4bb48f30b4ecd2fb1626181cb60d28081fc9d332b4ebefa.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "315433f37d34dc915912ea643e7ac277d3c7946de26c57d312057cd62a35e375", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "7cf3314f8e20856e20a25c41b973fb8526ac386fbdd6857046ee3ae555ac5cd6", "generation_seconds_batch": 22.108859014930204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.403374671936035, "target_z": 0.5117187094886272, "ce": 6.38003396987915, "genuineness": 4.582659721374512, "identity": 0.5512036085128784, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/315433f37d34dc915912ea643e7ac277d3c7946de26c57d312057cd62a35e375.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "3164ecf8c6258c53f84b58f145df98f6c0caf592e5e4e123834b5eb7b7110f0e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "6d8402949b4c8fe471cc45141fde468751d55ed9dcfe27c83aa8e6df3bbb1e15", "generation_seconds_batch": 26.342725471011363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9031970500946045, "target_z": -0.6640625401464055, "ce": 6.141753673553467, "genuineness": 4.632867813110352, "identity": 0.6666131019592285, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3164ecf8c6258c53f84b58f145df98f6c0caf592e5e4e123834b5eb7b7110f0e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "316b80f3ad25faf9f1656b9ed10afd70b7f9289440fe79dc4a274858612cb994", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "3b75d6e360f5466dacd3ce11ed26b2c6cd809918780e9b0883731a455d9c42df", "generation_seconds_batch": 26.76233570894692, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1655683517456055, "target_z": -1.2421874102180386, "ce": 6.260893821716309, "genuineness": 4.570107460021973, "identity": 0.5763139724731445, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/316b80f3ad25faf9f1656b9ed10afd70b7f9289440fe79dc4a274858612cb994.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "3178d41955b9183bf916a0a1fc5c4b514ee32ac36ac946a5606945f805547d73", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5e41d2c6da953f9a135128c2c1b64e16c46b373c40c98ed06967ea4be279a33e", "generation_seconds_batch": 22.288060329970904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.876852989196777, "target_z": 0.882812718250459, "ce": 5.847874641418457, "genuineness": 4.406930446624756, "identity": 0.3935268819332123, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3178d41955b9183bf916a0a1fc5c4b514ee32ac36ac946a5606945f805547d73.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "31ec0989ff536a41770a9ef24aedbe6dc45ce9020de206146a905455595e7af4", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "98ee34bf0f4405fd9f62b4706447cfbddef165b9d2072017b8fc9536399fd909", "generation_seconds_batch": 23.73444573301822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8199377059936523, "target_z": 0.054443360948921576, "ce": 6.0305562019348145, "genuineness": 4.720732688903809, "identity": 0.771579384803772, "hypothesis": "J'ai trouvé la 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_CASU_reg/31ec0989ff536a41770a9ef24aedbe6dc45ce9020de206146a905455595e7af4.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "321e3f70f797dd715a70e0bd8e9df2ec88767ca63abbedb0d25c078392f8eb84", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "a61edc6b5f861a8ee9719960709e74e5f77cc2cba6d3928e647f011349464c60", "generation_seconds_batch": 22.229210576973855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7567033767700195, "target_z": 0.004882737225489682, "ce": 6.149696350097656, "genuineness": 4.557555198669434, "identity": 0.4399804472923279, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/321e3f70f797dd715a70e0bd8e9df2ec88767ca63abbedb0d25c078392f8eb84.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "323c966cd45d1f28c4b6bc48b13a3840fc16ebd494ca88400496543941cdf809", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "634d46b5f3ecb159c2c6c083a0078dd676ab01556e691de8b90509de0267ebd1", "generation_seconds_batch": 21.91535487805959, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5953471660614014, "target_z": -0.12158202554738558, "ce": 5.736677646636963, "genuineness": 4.6203155517578125, "identity": 0.5972822308540344, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Manthe auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/323c966cd45d1f28c4b6bc48b13a3840fc16ebd494ca88400496543941cdf809.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "323d5dd9d047fb4f54d1e24e95d89578dd69793bad621f4212e891a15d91a3db", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "1de231f62fe28ed4b0a162fbf3e18949aa0b2ddeefd1f3c17422a079fec23bf2", "generation_seconds_batch": 19.52921697194688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3948311805725098, "target_z": -1.0625000642342488, "ce": 6.006728172302246, "genuineness": 4.469690799713135, "identity": 0.47143179178237915, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/323d5dd9d047fb4f54d1e24e95d89578dd69793bad621f4212e891a15d91a3db.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "32953bcdd5952e3692938555ab27bc2ded8c5ec41691a4786a3460e33fcac7ec", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6e67c35b54cfbbb5c305187ddf195983d4ad22ba2594b757de184e4289920095", "generation_seconds_batch": 22.606808132957667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.247091770172119, "target_z": -0.39453114744418233, "ce": 5.982900142669678, "genuineness": 4.457139015197754, "identity": 0.6155086159706116, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/32953bcdd5952e3692938555ab27bc2ded8c5ec41691a4786a3460e33fcac7ec.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "3346d1acc50ecc6bcacf63945f6785ae1f41e3289f1ac3afcd37a401f35fc2c4", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "24f1f47c65ddbc98722670e1ca4cd952538228e8d42718a30eec93d4c6c9d2fc", "generation_seconds_batch": 17.752890571951866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.980982542037964, "target_z": 0.180664080292157, "ce": 6.356205940246582, "genuineness": 4.570107460021973, "identity": 0.3987141251564026, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3346d1acc50ecc6bcacf63945f6785ae1f41e3289f1ac3afcd37a401f35fc2c4.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "3357b8b8a67f4309f96aae56651119927bd1cd058094eb6e3eeb94d0ce67fb01", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "7dbd8be3004ffde5b831d2c4ee35a63e6c122fbc6226e66e47fdd0d9a784e357", "generation_seconds_batch": 24.828695817966945, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.577280282974243, "target_z": -0.1357421503645749, "ce": 5.903473377227783, "genuineness": 4.413206577301025, "identity": 0.6579267978668213, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3357b8b8a67f4309f96aae56651119927bd1cd058094eb6e3eeb94d0ce67fb01.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "3386bbfb6dc47d0acd954d96d59126fa3047d213f26b64e424c229e3d3a75b6e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "c0662b299cbced31471ce4f221ee41bdc86f0462b6095deabc97823c2c861075", "generation_seconds_batch": 22.014224562095478, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6402029991149902, "target_z": -0.08642575693405212, "ce": 6.062326908111572, "genuineness": 4.469690799713135, "identity": 0.6669787168502808, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3386bbfb6dc47d0acd954d96d59126fa3047d213f26b64e424c229e3d3a75b6e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "33c2014481af2e3f12d01a9e89e8f478042dd3fc416f4244366200f82d6362b9", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "9745cacecf708e7746f7b77a71edbcaf9253d8cc4d81d87589be8463f5b5cffc", "generation_seconds_batch": 24.696139317005873, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.028330326080322, "target_z": 0.2177734437956863, "ce": 6.1894097328186035, "genuineness": 4.65797233581543, "identity": 0.5681244730949402, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/33c2014481af2e3f12d01a9e89e8f478042dd3fc416f4244366200f82d6362b9.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "33e2e8156257c41a6eac8362515a18af23314649d4063c3da2865edaee36c9c9", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "3460d80e58cb6bcdc7d151e71f99d729b670e42f154843da0f6f371c9ee03548", "generation_seconds_batch": 22.853858230053447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.562328338623047, "target_z": -0.1474609065690194, "ce": 6.292664527893066, "genuineness": 4.65797233581543, "identity": 0.6239256262779236, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/33e2e8156257c41a6eac8362515a18af23314649d4063c3da2865edaee36c9c9.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "768dd029fe8379441791e9622de140792fa8222bfe72f08e4796af9d0eda1e62", "generation_seconds_batch": 17.03093227208592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7337417602539062, "target_z": -0.7968750481756866, "ce": 6.078212261199951, "genuineness": 4.419482707977295, "identity": 0.7889223098754883, "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_CASU_reg/33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e.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": "33fd4ef997b7e012346fc6d15ff062245f9a36766b86d07c283d65b9342d962b", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b4cc7dcac94ecc4b37da19e0cbe1997547875f93c97a315d95951ed0ddf6367e", "generation_seconds_batch": 24.002198294969276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.483830690383911, "target_z": -0.20898432992653565, "ce": 5.824047088623047, "genuineness": 4.65797233581543, "identity": 0.7018731832504272, "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_CASU_reg/33fd4ef997b7e012346fc6d15ff062245f9a36766b86d07c283d65b9342d962b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "341a2d661282181f8ff48f84962b3b17f2ad641bf72f711fea4bc2ed911149e0", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "ede94da4f01cd1c851a2cec71c3dc9e102c71cc1a704cdb426bc390df03a9cd4", "generation_seconds_batch": 24.653218284947798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.094367980957031, "target_z": 0.26953120583895396, "ce": 5.403085231781006, "genuineness": 4.645420074462891, "identity": 0.7849075794219971, "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_CASU_reg/341a2d661282181f8ff48f84962b3b17f2ad641bf72f711fea4bc2ed911149e0.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "64d42afe4d9ab5c78a878ca3e0875b84786b98e7b6d9cade975d553e0db7d864", "generation_seconds_batch": 15.742170464014634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.195472240447998, "target_z": -1.2187498978091496, "ce": 6.356205940246582, "genuineness": 4.061748504638672, "identity": 0.3037251830101013, "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_CASU_reg/347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "3577d159809613ccc3b2e6aa267fcb34f84d0749181ad3b398d7d2069252b72a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "fc21150b94c166f6db0de147e05b3177ca14082cbdd6c6a7ed18ed40511852e5", "generation_seconds_batch": 22.722531103063375, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8788108825683594, "target_z": 0.10058588175124145, "ce": 6.316492557525635, "genuineness": 4.482243061065674, "identity": 0.5256557464599609, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3577d159809613ccc3b2e6aa267fcb34f84d0749181ad3b398d7d2069252b72a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "35b6e25f9c52292b3831e4d6fa4e9cf2c98d86920a25bcf24b927e3b060b444b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6619e4256c2ab5ff17c78a16aa847564815d0499b95d78df10cafd35ef669ab3", "generation_seconds_batch": 16.39809112693183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.069447994232178, "target_z": 0.2499998832104567, "ce": 5.879645347595215, "genuineness": 4.758388519287109, "identity": 0.6105407476425171, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/35b6e25f9c52292b3831e4d6fa4e9cf2c98d86920a25bcf24b927e3b060b444b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "35c19e9cb32c6d55d23dbf215a5b9a5a62e072204dc139c53bb2a0b92f331c4b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "3aa386a14cbf918ea1ef3193045cf16cd903e0156746c0c91908a6212cf592f1", "generation_seconds_batch": 23.804470272967592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.052004337310791, "target_z": 0.2363282189790856, "ce": 6.022613525390625, "genuineness": 4.344170093536377, "identity": 0.7261042594909668, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/35c19e9cb32c6d55d23dbf215a5b9a5a62e072204dc139c53bb2a0b92f331c4b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "35dfa6ddd190b02b895d281b9a132c6dd393dde04f4ad558071aa3e0b9d5c27d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "8d4d3eca8a6659264c3aa0e507542a3605ebc1f374c5b7649efda832ea24da2d", "generation_seconds_batch": 21.999566774000414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.281979560852051, "target_z": -0.3671874452549016, "ce": 6.356205940246582, "genuineness": 4.7081804275512695, "identity": 0.5550792217254639, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/35dfa6ddd190b02b895d281b9a132c6dd393dde04f4ad558071aa3e0b9d5c27d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "35e34e4033ddcc44bc0495bac844d5fa1666b1e4fb8b9a465fddc4aba5b6331a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "904d24c4990bc67a79fbdfa6f125992d5bf2ddaab6b51804eedb6fa124229c53", "generation_seconds_batch": 24.264147295965813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.464607000350952, "target_z": -1.007812472992418, "ce": 6.078212261199951, "genuineness": 4.300237655639648, "identity": 0.5285331010818481, "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_CASU_reg/35e34e4033ddcc44bc0495bac844d5fa1666b1e4fb8b9a465fddc4aba5b6331a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "3610a8745aa8033ec488fbf5efd0b895647e3344265b31294e255df555672329", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c9812bfebc04aafe71694caf244a3406106376912d72cb749ef7945f525860c8", "generation_seconds_batch": 25.366768251988105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2752156257629395, "target_z": -1.1562501138698047, "ce": 6.181467056274414, "genuineness": 4.13078498840332, "identity": 0.24911001324653625, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3610a8745aa8033ec488fbf5efd0b895647e3344265b31294e255df555672329.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "361e988ffb59ba25adc21938223457d4d6327629a6a5a7ac9eb8fc217d07d249", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aedba5d3e3af3ff7596eb36b98a7b0fb0b616fa93666f0127b920de0d2b1c8c7", "generation_seconds_batch": 21.302939523011446, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.396610975265503, "target_z": -0.27734377226300666, "ce": 6.086154937744141, "genuineness": 4.65797233581543, "identity": 0.6378798484802246, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/361e988ffb59ba25adc21938223457d4d6327629a6a5a7ac9eb8fc217d07d249.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "363fe867bf3fb15912c16c88fef639329f7f1bac8063afbe920678d73c0b1345", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "25993f61f7cd6969d70488d8811663d52861978460fe1ed2aaec27be3217cf26", "generation_seconds_batch": 16.82583533693105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.564286708831787, "target_z": -0.9296873693416985, "ce": 6.205295085906982, "genuineness": 4.570107460021973, "identity": 0.6078168153762817, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/363fe867bf3fb15912c16c88fef639329f7f1bac8063afbe920678d73c0b1345.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "36672246a74d1c3ddef68b87e8e7374374944ab4882f92957391f40087a498ea", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "deff6ea857e72c253a475a4fe7851576c6bf4f4c960cff9742c2acc57e99f21b", "generation_seconds_batch": 17.096238186000846, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.114304065704346, "target_z": 0.2851563386870594, "ce": 5.990842819213867, "genuineness": 4.821148872375488, "identity": 0.8281577825546265, "hypothesis": "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_CASU_reg/36672246a74d1c3ddef68b87e8e7374374944ab4882f92957391f40087a498ea.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "367824e91368a178acb45da1e534bab1b6931094b9d12e6dc2e716239d671541", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "04abd5a1082ab47b7502f903b5b556b32f3d71b91772b52804c918b8bca715ba", "generation_seconds_batch": 18.34246405295562, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.498070240020752, "target_z": 0.5859374364956859, "ce": 6.0146708488464355, "genuineness": 4.457139015197754, "identity": 0.667687714099884, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/367824e91368a178acb45da1e534bab1b6931094b9d12e6dc2e716239d671541.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": "3684374393be4196d76ea0ca6a47c3d5e556c3162e267429414779506782c65b", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4897d77322eb7f0f29be275f468fcc5cb96e6f2c5eda5cabcd276d48d89aeff2", "generation_seconds_batch": 21.13354949594941, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.023346424102783, "target_z": 0.21386725401529458, "ce": 5.80021858215332, "genuineness": 4.570107460021973, "identity": 0.7755228281021118, "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_CASU_reg/3684374393be4196d76ea0ca6a47c3d5e556c3162e267429414779506782c65b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "36befb6de930a23e9604147a535060d0dfde07a4d97786a3a0c1b609d92d3031", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9957708eafdf4edd7cb3818e94c5ccd7c4a3b2992d6d0a2643a6ae8003f7b369", "generation_seconds_batch": 19.043950177030638, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8383162021636963, "target_z": 0.06884771605902001, "ce": 5.959072113037109, "genuineness": 4.7081804275512695, "identity": 0.7236254811286926, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/36befb6de930a23e9604147a535060d0dfde07a4d97786a3a0c1b609d92d3031.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "36f16f1c7b0f4dadea68d4f1e2902cab69008795b5eaabfc6180d297ba78fe77", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "a95ecd5eab24cf9f02740c4140ec2fe47951293be2720c40b2f8cde4811e1fe9", "generation_seconds_batch": 24.704312894959003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8900249004364014, "target_z": 0.10937499562039213, "ce": 5.911416053771973, "genuineness": 4.570107460021973, "identity": 0.7266398072242737, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Y dime cuándo estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/36f16f1c7b0f4dadea68d4f1e2902cab69008795b5eaabfc6180d297ba78fe77.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "37062356514e2688f864eb6a16a965ad63eae36eba794a1320474c38e5c80187", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "2fec53a2131dc854b0adf1c96ba3b2df57b94d8018a253fa2b44d6af134bd826", "generation_seconds_batch": 23.804470272967592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.503054141998291, "target_z": 0.5898436262760776, "ce": 5.80021858215332, "genuineness": 4.457139015197754, "identity": 0.5978062152862549, "hypothesis": "De deur staan nog open. Ik heb je jas op de stoel gelegd. Wanneer je klaar bent.", "wer": 0.15789473684210525, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/37062356514e2688f864eb6a16a965ad63eae36eba794a1320474c38e5c80187.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "387de74ba0d37f31c5f3acfc3d7fe1b9b7374487f51267fcbe0bc1fb5ce0844b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "44ba128e78574e6f9b1d2638d32f8d1381d3bc545e74ba85b0988b039706524a", "generation_seconds_batch": 23.536171647021547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.657557487487793, "target_z": 0.7109373781009142, "ce": 6.109982967376709, "genuineness": 4.607763767242432, "identity": 0.3423730731010437, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/387de74ba0d37f31c5f3acfc3d7fe1b9b7374487f51267fcbe0bc1fb5ce0844b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "39e5d24dd4109d8b1d4161df7fcecc9fa78f98be3ca02458fa39d8d726fbbae7", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "658d3100c2ab989bb12420ee28130aed646b6fd37435c3c1d4daffe436235eed", "generation_seconds_batch": 22.64155014802236, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.311171054840088, "target_z": 0.43945307737176437, "ce": 5.7684478759765625, "genuineness": 4.2751336097717285, "identity": 0.30196335911750793, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen samen kijken.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/39e5d24dd4109d8b1d4161df7fcecc9fa78f98be3ca02458fa39d8d726fbbae7.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": "3a2bf8e8f0c43a2468a9d5ebb2395431a857e5a1093a3d52a29b4a0245a15a21", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e6a122a91b1ed943fbb508432fdff267073b92e9cd6f1b4b64fd0366be58f5e7", "generation_seconds_batch": 16.747511810041033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1591596603393555, "target_z": 0.32031242043712366, "ce": 6.022613525390625, "genuineness": 4.65797233581543, "identity": 0.8214558362960815, "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_CASU_reg/3a2bf8e8f0c43a2468a9d5ebb2395431a857e5a1093a3d52a29b4a0245a15a21.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "3a9ed4eeb25a50d3cddaf5754d430c2817d682b1e26171e5cb44612456ad9126", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "c1b55c650489732801303d4feb8af496d890c3c3933e3c59f80cec854f1e7f32", "generation_seconds_batch": 22.25931234296877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.071939945220947, "target_z": 0.2519529781006526, "ce": 5.744620323181152, "genuineness": 4.300237655639648, "identity": 0.2838633358478546, "hypothesis": "De door staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3a9ed4eeb25a50d3cddaf5754d430c2817d682b1e26171e5cb44612456ad9126.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": "3b08b6d12f6c4687d925a79eade25da41699503eb707e424109a799a1ddf5b57", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3e213235f06b64e58c051358dffde7c86d7156f4ea661d311761f8a186a93c4a", "generation_seconds_batch": 16.664834236027673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4551727771759033, "target_z": -0.23144529489032667, "ce": 5.935244083404541, "genuineness": 4.65797233581543, "identity": 0.6592179536819458, "hypothesis": "Les clés sont sur la table de la cuisine j'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3b08b6d12f6c4687d925a79eade25da41699503eb707e424109a799a1ddf5b57.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "3b388a58f4ad131ac373fbd006e4df120b8f6038525ec20aa58a306d8e473301", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "fd3053f57e7e256a037e41207f8dfb11123a9795a41410660038b4cd80b5e81a", "generation_seconds_batch": 22.563863244955428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7529654502868652, "target_z": 0.0019530948901958728, "ce": 6.141753673553467, "genuineness": 4.432034492492676, "identity": 0.446262001991272, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3b388a58f4ad131ac373fbd006e4df120b8f6038525ec20aa58a306d8e473301.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "3b4846f81337e25d7b8bceb921c6e418f8efaa673a4d024ed9784cd1fedd4071", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "b4408cb547759c7b4b08d8a3e6fe2ee09af178265bef153dd04c0f700d2c6cff", "generation_seconds_batch": 19.21011547197122, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7118477821350098, "target_z": -0.03027334452457453, "ce": 6.2767791748046875, "genuineness": 4.457139015197754, "identity": 0.3471881151199341, "hypothesis": "I found the little box. It was beside the old s 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_CASU_reg/3b4846f81337e25d7b8bceb921c6e418f8efaa673a4d024ed9784cd1fedd4071.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "3b6d99252f5a82e10a70d071bdf1c7ad5d44bab99d8da94e451b70d1572d5344", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "500d759a99f95e0d549dacd0975187ae054e7c0d5e49a0829f43950d16a4887d", "generation_seconds_batch": 24.70952018199023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8109042644500732, "target_z": 0.04736329854032691, "ce": 6.086154937744141, "genuineness": 4.6203155517578125, "identity": 0.6227103471755981, "hypothesis": "", "wer": 1.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3b6d99252f5a82e10a70d071bdf1c7ad5d44bab99d8da94e451b70d1572d5344.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "3ba384434c2797a1574c90c7e1fdd945f23745c1bf6626bc2fbe2b0529750232", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "2b8ae84dd194abd39cf50e2041ee23ec33d722c8adfed32451928d811643c7be", "generation_seconds_batch": 16.1072015799582, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9018619060516357, "target_z": 0.11865238321209177, "ce": 5.9431867599487305, "genuineness": 4.545003414154053, "identity": 0.6805742979049683, "hypothesis": "Ich hab die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen?", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3ba384434c2797a1574c90c7e1fdd945f23745c1bf6626bc2fbe2b0529750232.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "3a3094f678026483f4f49af6d932788b6945ec99a34632644b212f9cbe43c4fc", "generation_seconds_batch": 16.936040572007187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.255280017852783, "target_z": -1.1718748729913717, "ce": 6.252951145172119, "genuineness": 4.545003414154053, "identity": 0.5893346071243286, "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_CASU_reg/3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa.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": "3c8ed08599ae418b015db31a87b0f62516af80090548ac01dced238b7d8486d6", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ef4b7f66dbf1ee510da01ed391c6a73d0656af2210f16edcc2bce585b6d6cf9", "generation_seconds_batch": 15.70588179002516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5199644565582275, "target_z": -0.180664080292157, "ce": 5.839932441711426, "genuineness": 4.65797233581543, "identity": 0.5880379676818848, "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_CASU_reg/3c8ed08599ae418b015db31a87b0f62516af80090548ac01dced238b7d8486d6.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "3dd67a965da055edd54fe8351b4276b530683be56a800d6d4533338c93324292", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "729e6b8a50ddcf233bb24dd33d38780ff97dce214432acbe790280049f56accc", "generation_seconds_batch": 23.25060064601712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.446450710296631, "target_z": -0.23828131386928147, "ce": 6.260893821716309, "genuineness": 4.432034492492676, "identity": 0.3009243607521057, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3dd67a965da055edd54fe8351b4276b530683be56a800d6d4533338c93324292.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "3e0d25fa3cfd0850cfe3fd4c8b020e62315d23ab6558deb6dbf5ca92e0aaa984", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b089f2d08dbf01978bc5d318da7fb98652608595d30858534d6383a1ddbe1fd4", "generation_seconds_batch": 18.835496819927357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4925525188446045, "target_z": -0.2021484978108501, "ce": 6.24500846862793, "genuineness": 4.507347106933594, "identity": 0.3565271496772766, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3e0d25fa3cfd0850cfe3fd4c8b020e62315d23ab6558deb6dbf5ca92e0aaa984.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "3e29142b24c2f640bc3be414322ab9461f6a453b9bcf02caadb48b68dc136ee3", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "e16c8e8d7bb760d6a322918b447d40e254b90af3967d730c64fd5274be8a2cec", "generation_seconds_batch": 22.361798034980893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5984621047973633, "target_z": -0.11914065693464074, "ce": 5.871703147888184, "genuineness": 4.281409740447998, "identity": 0.6680926084518433, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3e29142b24c2f640bc3be414322ab9461f6a453b9bcf02caadb48b68dc136ee3.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "3e529ef7c07119e680d83bc917f63b28a1a883bca959340e6f12fbe8cdb37588", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "3e3504819c4d9943e603fe4df600b9601144e1028be7e4106db5ba08515c6fbb", "generation_seconds_batch": 23.682568341959268, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6034460067749023, "target_z": -0.115234467154249, "ce": 6.38797664642334, "genuineness": 4.532451152801514, "identity": 0.6584447622299194, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3e529ef7c07119e680d83bc917f63b28a1a883bca959340e6f12fbe8cdb37588.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "3ebc4a59afa131d8889a8b70f8b424c84cc91e19c819ca8e7cce82929dfc675b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "37bb8940d36fbd43aaf8da88c56a4bf77408573437eaeafc22e7a57ccc7ee566", "generation_seconds_batch": 17.30407881201245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.063218116760254, "target_z": 0.24511714598496703, "ce": 5.696964263916016, "genuineness": 4.582659721374512, "identity": 0.18624529242515564, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3ebc4a59afa131d8889a8b70f8b424c84cc91e19c819ca8e7cce82929dfc675b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "3ee13c720a46cdf3d724d9dd78ebc74112b13098d1d93a7be1ded2e5bfc074fb", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "ea511a4f5447117eadc03ec94050ed1abda75b95cb7345bb665e480f7aec928b", "generation_seconds_batch": 16.738397576031275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.723684549331665, "target_z": -0.02099614379614413, "ce": 5.498397350311279, "genuineness": 4.507347106933594, "identity": 0.6542493104934692, "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_CASU_reg/3ee13c720a46cdf3d724d9dd78ebc74112b13098d1d93a7be1ded2e5bfc074fb.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "3efeeb0ed7ae5921c634ac98351fb5ecafc45c9141d37d0a144603b78c681ce3", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d4fd2b7db6e441adca0bfa8a4ba8d65a4445aed3f17fc4cd053eb621dd87bfcd", "generation_seconds_batch": 17.064507020055316, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4895267486572266, "target_z": -0.9882813372271901, "ce": 6.316492557525635, "genuineness": 4.105680465698242, "identity": 0.2599249482154846, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3efeeb0ed7ae5921c634ac98351fb5ecafc45c9141d37d0a144603b78c681ce3.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "3fbc0665b2613b50d09f4b17c7c588398e7c946d2f22cf20503abba4872cc4a8", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4c893883a60d412071185b97aad3d5f2be8f91d61bca192f95649e11c7b789ca", "generation_seconds_batch": 21.09516625793185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.802805185317993, "target_z": 0.04101555328392108, "ce": 6.141753673553467, "genuineness": 4.3881025314331055, "identity": 0.2650817632675171, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/3fbc0665b2613b50d09f4b17c7c588398e7c946d2f22cf20503abba4872cc4a8.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": "3fe93653d09e2494a77b2cf8af10efa566b85cc7811c4b45c38d67103743a93e", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c7939675888e1a0173504f839b2b2a52fc28630c327801d96959a3da03797966", "generation_seconds_batch": 15.38395607098937, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.134239673614502, "target_z": 0.3007810978086264, "ce": 5.959072113037109, "genuineness": 4.507347106933594, "identity": 0.8882536888122559, "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_CASU_reg/3fe93653d09e2494a77b2cf8af10efa566b85cc7811c4b45c38d67103743a93e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "409d167e71684313702b247004f750d468b8fbaae6a8caf5f3a1deae15032ede", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "76f0f093b5761149acfd39d6c6d62f5dd8709383b3b97681ec9e4e29f6880c09", "generation_seconds_batch": 16.869817582075484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1025562286376953, "target_z": -0.5078125197082354, "ce": 6.022613525390625, "genuineness": 4.507347106933594, "identity": 0.5344160795211792, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/409d167e71684313702b247004f750d468b8fbaae6a8caf5f3a1deae15032ede.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "4125771ea6a2bd42447917c35ed5307a953d75dfaa4563f5f267ea2a0a13c622", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "42542ec670d40aed5aa78f48fcb2874f35b54f7935da0a56327178e2861a6d66", "generation_seconds_batch": 16.58148399204947, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.953570604324341, "target_z": 0.15917966277346388, "ce": 6.284721851348877, "genuineness": 4.3881025314331055, "identity": 0.38628530502319336, "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_CASU_reg/4125771ea6a2bd42447917c35ed5307a953d75dfaa4563f5f267ea2a0a13c622.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": "41beb2902c2a9c0e2fb886d858aedfb47309d2e5f1c5084d0f4f1c9e6aa96ec3", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The keys are on the kitchen table.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8f202b7befc32616d156ca1235991e867e3631d7b82d2b1d5273a2d0369d516b", "generation_seconds_batch": 17.144701304961927, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.308679103851318, "target_z": 0.4374999824815685, "ce": 6.1894097328186035, "genuineness": 4.494794845581055, "identity": 0.49550050497055054, "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_CASU_reg/41beb2902c2a9c0e2fb886d858aedfb47309d2e5f1c5084d0f4f1c9e6aa96ec3.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "41d6c8e07a37e668bf8c8e8eb7bef446e1d9f13cdb6094604c72d46919577f80", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "cf9fe19cfdc6482f4b8986964fa0d863ba451dbbbfaff9f23541409606176c87", "generation_seconds_batch": 21.289421800989658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6490142345428467, "target_z": -0.8632812087586925, "ce": 6.022613525390625, "genuineness": 4.532451152801514, "identity": 0.6099326610565186, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/41d6c8e07a37e668bf8c8e8eb7bef446e1d9f13cdb6094604c72d46919577f80.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": "41e8e2d1d93c614cfbe794f0cb85f025ee55115af8dd70f2840629a5d9224c9e", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5b3d98bc9efa2f095a5a8036e85b81c9b1f178915a748357d10cd5644549737", "generation_seconds_batch": 17.161306614987552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7741475105285645, "target_z": 0.01855477518339929, "ce": 5.700935363769531, "genuineness": 4.65797233581543, "identity": 0.8204154968261719, "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_CASU_reg/41e8e2d1d93c614cfbe794f0cb85f025ee55115af8dd70f2840629a5d9224c9e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "422f11e0f5e695b253495338e3860b03d73e3623b7bf51363833b510991fc0b3", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "aaa81ce742a82098a7b24c3ce9df5e6f0e4ed7912af1216ee96e7d138c2f0996", "generation_seconds_batch": 26.84206254093442, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8769419193267822, "target_z": 0.09912106058359454, "ce": 6.046441555023193, "genuineness": 4.532451152801514, "identity": 0.4917218089103699, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/422f11e0f5e695b253495338e3860b03d73e3623b7bf51363833b510991fc0b3.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "4246da5bcd8da3fa41d09727605a130c0d7e0177007a1ca36efa7af17032a779", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "0f7b8f66817489863f26e8c0f105d2eebb7b8195d2d5d40afb8c22e502eba0e8", "generation_seconds_batch": 21.743965385016054, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5279741287231445, "target_z": 0.6093749489045748, "ce": 5.824047088623047, "genuineness": 4.507347106933594, "identity": 0.3916334807872772, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/4246da5bcd8da3fa41d09727605a130c0d7e0177007a1ca36efa7af17032a779.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "43124bad44da5519fafd369c6c62eb226cd77e94a2904d901c86bbc316c38857", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "7847e5be599ae7269970125a6d3a2e0cbdb1877c6e3ba3c71461a6dec7651fd1", "generation_seconds_batch": 22.877861893968657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8982131481170654, "target_z": -0.6679687299267972, "ce": 5.990842819213867, "genuineness": 4.457139015197754, "identity": 0.5060768723487854, "hypothesis": "Die Tür ist noch offen. Ich habe dein Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/43124bad44da5519fafd369c6c62eb226cd77e94a2904d901c86bbc316c38857.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "0feb2a8fc625bb2bc0d93cc585d9ba9a2a9e4594d4a8c86c6b8a2f603d05ce52", "generation_seconds_batch": 17.03093227208592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5842223167419434, "target_z": -0.9140626102201315, "ce": 6.125868320465088, "genuineness": 4.432034492492676, "identity": 0.6487776041030884, "hypothesis": "De weg gaat linksaf naar de brug, loopt verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "437d1845ecc899aa2eaf13da575cd4f64ad04300ddeb793dc0e97e3d8d61f1f8", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0cdbd0f0443840470b74892abd41c85ff12c5caf474dd0494631087a965d386e", "generation_seconds_batch": 22.1925501950318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.718700647354126, "target_z": -0.024902333576535877, "ce": 5.9431867599487305, "genuineness": 4.7081804275512695, "identity": 0.7055954933166504, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/437d1845ecc899aa2eaf13da575cd4f64ad04300ddeb793dc0e97e3d8d61f1f8.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "439fab68ee41ebdf2fcf5069c5b51c70bb79cdc1aa0f88053271ec251635ac88", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "298673be37904efd259279d8e5d24adcb24494e203d0f6435c2e2bb07259bf21", "generation_seconds_batch": 23.262227679020725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7088217735290527, "target_z": -0.8164063708041839, "ce": 6.046441555023193, "genuineness": 4.093128204345703, "identity": 0.5055093765258789, "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_CASU_reg/439fab68ee41ebdf2fcf5069c5b51c70bb79cdc1aa0f88053271ec251635ac88.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "0e6d11c81ee3211eb14e3500ea39a587fceec4b879e69c446415e63a4b564a0e", "generation_seconds_batch": 16.36459890601691, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7037487030029297, "target_z": -0.036621089780980366, "ce": 6.173524379730225, "genuineness": 4.65797233581543, "identity": 0.5983486175537109, "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_CASU_reg/43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "43f93cfd00630b98b0dc36994e194332335cb0536af20bb076b1bb6e041183f7", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "732454f45b90ff5cf393fa5c2455c3528d3235953b81e81c274cc3f0f4461b26", "generation_seconds_batch": 21.53266403498128, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.241395473480225, "target_z": 0.38476567299320297, "ce": 6.0305562019348145, "genuineness": 4.507347106933594, "identity": 0.6192530393600464, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/43f93cfd00630b98b0dc36994e194332335cb0536af20bb076b1bb6e041183f7.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "444d2fee4fce53e0c467dc9cf3a465f001e527be90c2f396ba87c16f02e28a84", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a57fa3ea59d3a1a037d6c440620d3dff5dfc757944d442953b59f0fa09eb6218", "generation_seconds_batch": 23.377249875920825, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.787140846252441, "target_z": 0.8124998072972536, "ce": 5.990842819213867, "genuineness": 4.833701133728027, "identity": 0.6490366458892822, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/444d2fee4fce53e0c467dc9cf3a465f001e527be90c2f396ba87c16f02e28a84.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "448f7005bc1333457b4785fdbc5b4877aa0c5821d3a9dce922985db0d210ce03", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "20bdf908e76549c5e0e7134addda140ea5c261419e538202544f61c9d80accab", "generation_seconds_batch": 19.043950177030638, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.591609239578247, "target_z": -0.1245116678826794, "ce": 6.022613525390625, "genuineness": 4.582659721374512, "identity": 0.7536238431930542, "hypothesis": "La porte est encore ouverte , j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/448f7005bc1333457b4785fdbc5b4877aa0c5821d3a9dce922985db0d210ce03.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "4546c98ab4baf87b158331cf4a80a4b5a5e4793706a3e4d66396d7ea6213b608", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "ee696c0d43e8e32f6b66c4c1f14e917625d36b57cdeb6e9ca6141f4c46b874e4", "generation_seconds_batch": 24.5633657600265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7985334396362305, "target_z": -0.7460938335775169, "ce": 5.95112943649292, "genuineness": 4.469690799713135, "identity": 0.6803966760635376, "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_CASU_reg/4546c98ab4baf87b158331cf4a80a4b5a5e4793706a3e4d66396d7ea6213b608.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "45745179206c739458b1930ebe6ed181c1927924821a92e47756515b5f957140", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "5108ce87ed74b9eda550010061b84fa7b24c0d8e5315064ea7a41b73873d0079", "generation_seconds_batch": 24.703866609022953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7940833568573, "target_z": 0.034179721168235526, "ce": 6.340320587158203, "genuineness": 4.444586753845215, "identity": 0.5286802053451538, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/45745179206c739458b1930ebe6ed181c1927924821a92e47756515b5f957140.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "61f1ad4bacf411b1fbdf15c0f3eb082773e884c3545be44a4aae0bf8e5801389", "generation_seconds_batch": 18.939843983971514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0559206008911133, "target_z": -1.328125080292811, "ce": 6.4912309646606445, "genuineness": 4.18726921081543, "identity": 0.5834434032440186, "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_CASU_reg/45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "4616f48f7e1928342d0f3ef3630ed6b39449b6e73e43eec60339680b238cefd4", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "f5a06222f0e2f14f494cf15c81c5ede5ed53d06ebaad33e18fbcd36088f51e73", "generation_seconds_batch": 23.26663729501888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.096859931945801, "target_z": 0.2714843007291498, "ce": 5.967014789581299, "genuineness": 4.482243061065674, "identity": 0.1340036243200302, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/4616f48f7e1928342d0f3ef3630ed6b39449b6e73e43eec60339680b238cefd4.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": "462364b14bb9f1a6f29789ee30d641a7dc3005b99ee3a49547a5a0222f2df494", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Your message arrived after I had left.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1015a50cf9feb3171ca830283bd7b302636a8870dd934748d432916a15d29f86", "generation_seconds_batch": 21.428442695993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9784905910491943, "target_z": 0.17871098540196112, "ce": 6.141753673553467, "genuineness": 4.369274139404297, "identity": 0.2992382347583771, "hypothesis": "Your m message arrived after I had left. I read it twice before putting the phone away. Let me know and we can talk.", "wer": 0.08695652173913043, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/462364b14bb9f1a6f29789ee30d641a7dc3005b99ee3a49547a5a0222f2df494.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "462ddfd357d2322273fba0e11a748f1a88318465d5d81172ed6496df6337854d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "295f1964bd3e2c4b75566d7f07e05307087fc5dca2d1c54395fed648ccb5b48a", "generation_seconds_batch": 23.910638179047965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.928117036819458, "target_z": -0.6445312175179083, "ce": 6.09409761428833, "genuineness": 4.65797233581543, "identity": 0.3043171763420105, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/462ddfd357d2322273fba0e11a748f1a88318465d5d81172ed6496df6337854d.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": "4665d70d27bc4a09bb1781031d4237b37d78b0d22c152102850d633a7e57b976", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e6e393441b36ae650338683fb84e1bd5844c85c34df8a57e10ff806b78f0b26e", "generation_seconds_batch": 15.70588179002516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7710323333740234, "target_z": 0.0161132197073852, "ce": 6.022613525390625, "genuineness": 4.557555198669434, "identity": 0.7478228211402893, "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_CASU_reg/4665d70d27bc4a09bb1781031d4237b37d78b0d22c152102850d633a7e57b976.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "468a0c3660d597643d29d3fa1cb562409c2d7fab6583136605b84846823f8726", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ddbd034bcad3f176f3ae048ef47bb09834ba3ecf1d6d6bebd1bb37671b412310", "generation_seconds_batch": 21.91535487805959, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1947598457336426, "target_z": -0.4355468875913726, "ce": 5.728734970092773, "genuineness": 4.720732688903809, "identity": 0.47901296615600586, "hypothesis": "Tier ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/468a0c3660d597643d29d3fa1cb562409c2d7fab6583136605b84846823f8726.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "46d5e6cc4fbb79b6a495ccbb1cf64d2c2a759e4978c76ea37457e13081d44c42", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "04aae13376248a407212d6e29f9ed3610d335f95bf988e88a87164a13f8ef416", "generation_seconds_batch": 18.809754970949143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7467355728149414, "target_z": -0.0029296423352938093, "ce": 5.967014789581299, "genuineness": 4.381826400756836, "identity": 0.8010265827178955, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es Précipréia.", "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_CASU_reg/46d5e6cc4fbb79b6a495ccbb1cf64d2c2a759e4978c76ea37457e13081d44c42.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "476cb59f0834feb98d921872dd996eb6459c946d0783425393c62d520dd5bf93", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d24447720032790b6c92302bb455b71bb3af4c6db97deed648de15b7dfcba6f7", "generation_seconds_batch": 19.496425554039888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.54986834526062, "target_z": -0.157226567883268, "ce": 6.046441555023193, "genuineness": 4.369274139404297, "identity": 0.46376103162765503, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/476cb59f0834feb98d921872dd996eb6459c946d0783425393c62d520dd5bf93.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "477edc7ae2b026daec1db0b65dc746e5de8fcd4cf1fb0f6de4232afa0d2cbb14", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "93fdc6d626cf2a7bd41f2aa43980ac77a1fb201e53aa7b057b022f1247904ff5", "generation_seconds_batch": 17.84299272496719, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4800925254821777, "target_z": -0.2119141591250987, "ce": 6.173524379730225, "genuineness": 4.507347106933594, "identity": 0.4699709117412567, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/477edc7ae2b026daec1db0b65dc746e5de8fcd4cf1fb0f6de4232afa0d2cbb14.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "ec7785060d912809b11dc98e143782cc1aa6b67f94476649bc910d3dcafb1bcf", "generation_seconds_batch": 16.51525559008587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.92313289642334, "target_z": -0.6484375941615693, "ce": 6.23706579208374, "genuineness": 4.532451152801514, "identity": 0.663874626159668, "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_CASU_reg/47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d.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": "47fe4ae859d016f9bea83037378e68fdd5832b05a92778d5d143809a3c1649de", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad967915548a68135e7ed9e71b3fdd2fb8e422642b753493cfb29f7ee81d933f", "generation_seconds_batch": 22.26796659291722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.733963966369629, "target_z": -0.012939533942451531, "ce": 5.998785495758057, "genuineness": 4.7081804275512695, "identity": 0.5616233944892883, "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_CASU_reg/47fe4ae859d016f9bea83037378e68fdd5832b05a92778d5d143809a3c1649de.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "48946fee5e1df8358b51e507bc38c09c3df9d491fb032e2e50e1502c0b53b0d5", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "924e20d06421ecdd815145f2b7d67238f728c2b9b3904dd392e302a62d7bc83d", "generation_seconds_batch": 16.865278732031584, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3442792892456055, "target_z": -0.31835932554692775, "ce": 5.7684478759765625, "genuineness": 4.419482707977295, "identity": 0.6755046844482422, "hypothesis": "La puerta sigue abierta. He dejado tu ahrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/48946fee5e1df8358b51e507bc38c09c3df9d491fb032e2e50e1502c0b53b0d5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "48a87835f41e8f378b0c3f16ebaa58e9e0b337c997f1fb045645307e8761d30c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "027b9851ba08d6c462be232e66c6110c9fa0d273d05d3806b87d8e3edd6c3820", "generation_seconds_batch": 16.664915927918628, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6763367652893066, "target_z": -0.05810550729967347, "ce": 5.597680568695068, "genuineness": 4.532451152801514, "identity": 0.7423281669616699, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/48a87835f41e8f378b0c3f16ebaa58e9e0b337c997f1fb045645307e8761d30c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "49362b6808ac0ea506165a47db84cbaf872f35dcaa68cbfedfa9559d3d8e271f", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "68ef48c23a6295c6817603a2c0d58c7b348390f4c4b2bd398743b644945eb0f6", "generation_seconds_batch": 22.68132184201386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.967988967895508, "target_z": -0.6132811386849666, "ce": 6.0305562019348145, "genuineness": 4.2249250411987305, "identity": 0.5538488626480103, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/49362b6808ac0ea506165a47db84cbaf872f35dcaa68cbfedfa9559d3d8e271f.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "d9abf337fa9844a72f0248a5b88a7d28a0a987064d0661cf73ce886c23481203", "generation_seconds_batch": 16.44140391901601, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2652478218078613, "target_z": -1.164062493430588, "ce": 6.0305562019348145, "genuineness": 4.21237325668335, "identity": 0.6542391777038574, "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_CASU_reg/49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "4a12a3df77d0619f3b9396e79ba36bc88f43dc66cb659f74c79ef7d642519692", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b39a7b5159bba2fe293e2c84e6b5e7df4d2663cf0c87a3c22e43a8a54e68c3ae", "generation_seconds_batch": 18.48932155710645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5037665367126465, "target_z": -0.1933593839416994, "ce": 6.054384231567383, "genuineness": 4.720732688903809, "identity": 0.7235244512557983, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/4a12a3df77d0619f3b9396e79ba36bc88f43dc66cb659f74c79ef7d642519692.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "4b15eb0f3839bbde0ac4393ff232a717bf35136d4e1c7df5742f34046faf3173", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "c280bd50df875db3ac4306b1a869aceb4281e300bb75e95e2abc935914277f42", "generation_seconds_batch": 17.14279102499131, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.032780647277832, "target_z": -0.5624999240867968, "ce": 6.09409761428833, "genuineness": 4.758388519287109, "identity": 0.6841191053390503, "hypothesis": "Ich hab die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/4b15eb0f3839bbde0ac4393ff232a717bf35136d4e1c7df5742f34046faf3173.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "4bc5d82a345c43d1a98ef617021a676df4c68338654bf1e2c499561959a1741b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7eb5f590ecf98755084149e613c9add14a11b1fb88decc19ac08a4977ec4b8eb", "generation_seconds_batch": 22.65299829491414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.089384078979492, "target_z": 0.2656250160585622, "ce": 6.038498878479004, "genuineness": 4.632867813110352, "identity": 0.6693967580795288, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/4bc5d82a345c43d1a98ef617021a676df4c68338654bf1e2c499561959a1741b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "4c2cf534a8a51690fdfe493d90a1cd8db98af2a20cf150f1ce467a2ce309eefd", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "b93a3bd4aa7036b3944b3ec59f7b919a9890bae4ecfcb95683f6f5a39118ccb8", "generation_seconds_batch": 17.446140857995488, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.468166351318359, "target_z": 0.5624999240867968, "ce": 5.990842819213867, "genuineness": 4.375550270080566, "identity": 0.659995973110199, "hypothesis": "Ik heb het klein doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/4c2cf534a8a51690fdfe493d90a1cd8db98af2a20cf150f1ce467a2ce309eefd.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "4cb561ee791985ee31e6b8feaf839b0854f0c0a501c70093da44c0e6c85dab97", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c63c03a59dd8776fbf40e5126cb484098a5d2c83d79f3bb5cdfa31d96d64cbdf", "generation_seconds_batch": 22.151396099012345, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.166635513305664, "target_z": 0.32617170510771126, "ce": 5.6413655281066895, "genuineness": 4.507347106933594, "identity": 0.7100253105163574, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt ?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/4cb561ee791985ee31e6b8feaf839b0854f0c0a501c70093da44c0e6c85dab97.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "4cee097cbc5c7ef465d66fb2583a6741e684d2aeef282dddce8da27b163af885", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "eb877f493d0361e7e7c28077873196d85a3c259e658d68cefa9b009f7270782e", "generation_seconds_batch": 16.7550384439528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5099964141845703, "target_z": -0.18847664671620973, "ce": 5.8875885009765625, "genuineness": 4.720732688903809, "identity": 0.6399973034858704, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/4cee097cbc5c7ef465d66fb2583a6741e684d2aeef282dddce8da27b163af885.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "4d06bcfb7f7d740e26a543f9a45febffe4f6dc5cc47070637983a22d21e64761", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "0975dccda86c7139fda2e6302ed4f240fa95ab1c36080cfb70cecb9830b5f47b", "generation_seconds_batch": 16.795615051989444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.961580276489258, "target_z": 0.9492186919701957, "ce": 6.086154937744141, "genuineness": 4.595211505889893, "identity": 0.41804105043411255, "hypothesis": "I found the little box. It was beside the old window. We can look inside together,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/4d06bcfb7f7d740e26a543f9a45febffe4f6dc5cc47070637983a22d21e64761.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "4d0a7d1e2d7ccb40fc5d29eb933077f3c7565b8797f3b0e62eed646275700a2a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "da563a20a336159fd5fb228f0a6230819a722db53db6a0989030ada8274a4efc", "generation_seconds_batch": 23.1150081200758, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5941903591156006, "target_z": -0.9062500437960788, "ce": 6.09409761428833, "genuineness": 4.13078498840332, "identity": 0.6052021980285645, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/4d0a7d1e2d7ccb40fc5d29eb933077f3c7565b8797f3b0e62eed646275700a2a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "4e7d5c4efd13b73f4eb9123839f83e93e2d2b38c1c21b4af3857def5ac80160e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "b270fa4b3053d1a8996d30616f3dc40d0329d6ae14739e777e86cadc15970b1c", "generation_seconds_batch": 15.347349507035688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.556098222732544, "target_z": -0.15234383065777832, "ce": 5.792276382446289, "genuineness": 4.20609712600708, "identity": 0.21079280972480774, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/4e7d5c4efd13b73f4eb9123839f83e93e2d2b38c1c21b4af3857def5ac80160e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "4ed7195a956aff30dc34e527f78aa63db781c79ea53e89e3a43fe982f256a265", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "6ff426c022b0c704c4ff72ddcf3c5f1f2c3ff0b5b274f49f0ef41cd880aed65e", "generation_seconds_batch": 22.463291352032684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.522456407546997, "target_z": -0.17871098540196112, "ce": 5.935244083404541, "genuineness": 4.7081804275512695, "identity": 0.0967593789100647, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/4ed7195a956aff30dc34e527f78aa63db781c79ea53e89e3a43fe982f256a265.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": "4f8ccd4022ee808ffd0166ba220baa9c6167961bc25dd42141bd4d501d5c3d53", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Escribí el número en el sobre.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a16d2341f4bcb55b4adf2d8b65a062588d838687b0c1669f31183ad8a6b9649b", "generation_seconds_batch": 24.85774082201533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.238903522491455, "target_z": 0.38281257810300706, "ce": 6.038498878479004, "genuineness": 4.6705241203308105, "identity": 0.616104245185852, "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_CASU_reg/4f8ccd4022ee808ffd0166ba220baa9c6167961bc25dd42141bd4d501d5c3d53.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "4fb1cbd8656eab2a7e0a8948165cca41bacc4f4dad937d773edd0d2e86177720", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "04c4828c228a01516b6bd90b0683e0585a26383092ed48ecc45fb23423f72548", "generation_seconds_batch": 25.300991802942008, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9080917835235596, "target_z": 0.12353512043758146, "ce": 5.7128496170043945, "genuineness": 4.18099308013916, "identity": 0.6731911897659302, "hypothesis": "Die deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/4fb1cbd8656eab2a7e0a8948165cca41bacc4f4dad937d773edd0d2e86177720.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "4fbac777abd3398df6641b1f0071d7f1b2c34ab0a5e4f6931254a38bcf940488", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "104d5383479ea25aeb5c059ffb3358aa5c3f32ccd593015d0028188e9ad6c686", "generation_seconds_batch": 17.70816905994434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2620434761047363, "target_z": -0.38281257810300706, "ce": 6.046441555023193, "genuineness": 4.946669578552246, "identity": 0.8150781393051147, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/4fbac777abd3398df6641b1f0071d7f1b2c34ab0a5e4f6931254a38bcf940488.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "4fd4adc186d548baeb6e16448612841bda4548dfecd7ec230decfa3b6dbbd21e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1844a6fd1a91947f7f1e78f89888dfb4fd1c6f310041cd87242a92b30b0b5f22", "generation_seconds_batch": 19.071058742003515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1673479080200195, "target_z": -0.45703130511006573, "ce": 5.935244083404541, "genuineness": 4.356721878051758, "identity": 0.33673304319381714, "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_CASU_reg/4fd4adc186d548baeb6e16448612841bda4548dfecd7ec230decfa3b6dbbd21e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "4ffed1bf2d49fcc7eca3820beb771c31d13c181b5fe28ef007888bd4fbf6f44b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "5f9ea48b12f58b155ff223074d7123431e40027d0f2d984f6bcad85ab663ac9e", "generation_seconds_batch": 25.114051705924794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2271556854248047, "target_z": -0.4101562802922878, "ce": 6.1894097328186035, "genuineness": 4.607763767242432, "identity": 0.5170626640319824, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/4ffed1bf2d49fcc7eca3820beb771c31d13c181b5fe28ef007888bd4fbf6f44b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "504a1c9f7f155d635d133b73210137f93dc460420532efe6400887e8c6590fcc", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "4282dd3e9c8e09a27314052dc36e0f2ad587e267d8973818f6a3422e4c754704", "generation_seconds_batch": 23.645970265963115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.385931015014648, "target_z": 0.49804704525725607, "ce": 6.181467056274414, "genuineness": 4.381826400756836, "identity": 0.4386546313762665, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/504a1c9f7f155d635d133b73210137f93dc460420532efe6400887e8c6590fcc.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "51704f40b63529ca28a20113b2b20ce86e57321aec8dbc3502c81767c141bdc9", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3893dedd40097c79c69912061608ed6673889aa1e4bc0f959d23189229e733e8", "generation_seconds_batch": 22.909954982926138, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3218512535095215, "target_z": -0.3359375532852291, "ce": 6.09409761428833, "genuineness": 4.293961524963379, "identity": 0.38150709867477417, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/51704f40b63529ca28a20113b2b20ce86e57321aec8dbc3502c81767c141bdc9.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "51a35be8045f8bf46d4faa7a82bb7b5edf9125e775a85050cb2ca03d670ba436", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "60efab260dc033b7e7985bfbf9457317695a113d5d0e8940305c016c434c3d2b", "generation_seconds_batch": 22.766464699991047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.646432876586914, "target_z": -0.08154301970856244, "ce": 5.593709468841553, "genuineness": 4.607763767242432, "identity": 0.7370677590370178, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/51a35be8045f8bf46d4faa7a82bb7b5edf9125e775a85050cb2ca03d670ba436.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "51bee24e8ca9a8449df10e35f4b3b36c8e2297a0bc250a3ef408e86246ce1af3", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "38b5bc9176c2fda0cd2e44e1dd2515268f1a69284e1715774cbbb2efdc7331dd", "generation_seconds_batch": 23.262227679020725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.306899309158325, "target_z": -0.34765630948967363, "ce": 5.935244083404541, "genuineness": 4.394378662109375, "identity": 0.35784441232681274, "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_CASU_reg/51bee24e8ca9a8449df10e35f4b3b36c8e2297a0bc250a3ef408e86246ce1af3.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "51d543cd6437211ecbc55fb2fd560a253f5a28c274470839cd1bcea2bc2b8156", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "24579bd708ec7661cc5529632c89d826fdd0e1d03d496ed2def2662034986695", "generation_seconds_batch": 24.696139317005873, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.119287967681885, "target_z": 0.2890625284674512, "ce": 5.911416053771973, "genuineness": 4.7458367347717285, "identity": 0.7386519908905029, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/51d543cd6437211ecbc55fb2fd560a253f5a28c274470839cd1bcea2bc2b8156.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "521c5c72c59e3ae902a8892799212870cf02296dffde7862cdc5d99cd214e944", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "82d0f9a63554c0e363491722cc285a8557acefe97f759d7ad1175b6daf2bef7d", "generation_seconds_batch": 21.284669721964747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8465936183929443, "target_z": -1.4921874802917645, "ce": 6.38003396987915, "genuineness": 4.444586753845215, "identity": 0.6382665634155273, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/521c5c72c59e3ae902a8892799212870cf02296dffde7862cdc5d99cd214e944.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "530aba343b6715de1f04483bd1c26ee0115f9ab35b486124d30d0cbbe3bf726a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "068c60c4b560ecee6746d9fe63171fa330a4bdddd70d746b1980c830112223a0", "generation_seconds_batch": 22.499581065028906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3592312335968018, "target_z": -0.3066405693424833, "ce": 5.514282703399658, "genuineness": 4.319066047668457, "identity": 0.10533399879932404, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/530aba343b6715de1f04483bd1c26ee0115f9ab35b486124d30d0cbbe3bf726a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "535ae5d945008aa09d0dc7951ff797f4ad3808d4bfa3403a9ab45c92f73dcb4b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "9b3b01a5445136e906553274e37cc8c3968d9957c687d1add326790c72be9d52", "generation_seconds_batch": 22.606799453962594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.092588424682617, "target_z": -0.5156248992690189, "ce": 6.022613525390625, "genuineness": 4.65797233581543, "identity": 0.8011291027069092, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/535ae5d945008aa09d0dc7951ff797f4ad3808d4bfa3403a9ab45c92f73dcb4b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "537d6f84e51d1bababafd6b8c7fb117e3a3cf0c4aca5d8d34fb8531c1f730b09", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "2338bdeabf6c583360b47f9489c10c55165dd2c7f4ced75c371e6aca7053bc79", "generation_seconds_batch": 16.869817582075484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6440300941467285, "target_z": -0.8671875854023535, "ce": 6.117925643920898, "genuineness": 4.607763767242432, "identity": 0.6316232681274414, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Manthe auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/537d6f84e51d1bababafd6b8c7fb117e3a3cf0c4aca5d8d34fb8531c1f730b09.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "538cbcd5d36cd30712fa048afa3693f878e6f7b260b9903fb6ce2cee81e281e0", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "d314259b790a39cddbf53104fbcc04ff733dec2cec1fba9a60384eb157b62565", "generation_seconds_batch": 22.788602139102295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7224385738372803, "target_z": -0.02197269124124207, "ce": 6.086154937744141, "genuineness": 4.482243061065674, "identity": 0.800996720790863, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/538cbcd5d36cd30712fa048afa3693f878e6f7b260b9903fb6ce2cee81e281e0.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "53bcd858446aa00620ce2dca80f17da07ed7f8162172abc34ff9633b5d01f8c9", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "5564d536c27937ce4705a46619864448b836c35b63d449ad828080a495f9096f", "generation_seconds_batch": 17.193021383020096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5959701538085938, "target_z": -0.12109375182483662, "ce": 6.109982967376709, "genuineness": 4.532451152801514, "identity": 0.47632649540901184, "hypothesis": "Die Tür 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_CASU_reg/53bcd858446aa00620ce2dca80f17da07ed7f8162172abc34ff9633b5d01f8c9.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": "53dc30093d78304950824a6da51a17302b8e7c43a4426a3b2eb4fa0791e90b3a", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db47aaab788bfd8f5cc2dd9609f2ed9959545c00530e4eff9959abaf122375d5", "generation_seconds_batch": 17.190951670054346, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.763645648956299, "target_z": -0.7734375357667976, "ce": 6.1894097328186035, "genuineness": 4.457139015197754, "identity": 0.7084108591079712, "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_CASU_reg/53dc30093d78304950824a6da51a17302b8e7c43a4426a3b2eb4fa0791e90b3a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "5410aba4a551d1087dc398fbc1c4bd10a0fe070478b605b5b9105d39aef388f1", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "e8ae99bf231eec2c382efeebe9ef057d5f4632736a1946569007b289e75dfd4f", "generation_seconds_batch": 22.382014365051873, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9722604751586914, "target_z": 0.1738280613132022, "ce": 5.919358730316162, "genuineness": 4.68307638168335, "identity": 0.7133172750473022, "hypothesis": "La porte est encore ouverte : J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/5410aba4a551d1087dc398fbc1c4bd10a0fe070478b605b5b9105d39aef388f1.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "54284002ed45c810f35309646222eb403b91e4571413e94284087b5a73d78743", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "5614a50742d2acfc45df80eb6385e7a99e095988b19e4fa1b5078aebaaa9edc1", "generation_seconds_batch": 24.28665100398939, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.508038520812988, "target_z": 0.5937501897830079, "ce": 6.109982967376709, "genuineness": 4.758388519287109, "identity": 0.6803946495056152, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/54284002ed45c810f35309646222eb403b91e4571413e94284087b5a73d78743.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "543d1f073f2066bb08957781221f22833f39e12060ee578a2d43b2ad1ce880e7", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ffe520bb685f725a3f669311624e251d10569d344f7ca1519e5c473d1d35736a", "generation_seconds_batch": 21.43731519393623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3667070865631104, "target_z": -0.3007812846718957, "ce": 5.911416053771973, "genuineness": 4.494794845581055, "identity": 0.5579990744590759, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/543d1f073f2066bb08957781221f22833f39e12060ee578a2d43b2ad1ce880e7.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "9fd6bd05df255a771b6c5ec4fce6005eaaf4396a5e447222ab3a6a7db49a7101", "generation_seconds_batch": 18.939843983971514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.258484125137329, "target_z": -1.953124975182222, "ce": 6.268836498260498, "genuineness": 3.578493118286133, "identity": 0.5218927264213562, "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_CASU_reg/546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "550dfc9e85907969229ffa54134d92a3dba20f4ed73f4b59fb15d45a231892fe", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "ef953d791e93b045866409a813d9c61407b437a72aa80d59f3ef8bd38a1e8652", "generation_seconds_batch": 17.76293143199291, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.483830690383911, "target_z": -0.20898432992653565, "ce": 6.157639026641846, "genuineness": 4.306513786315918, "identity": 0.22569961845874786, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/550dfc9e85907969229ffa54134d92a3dba20f4ed73f4b59fb15d45a231892fe.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "557def5078ebb2c18c25c9f74b9a9c6b1f3dd6736c504840dba1ce90ef4b58a6", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "8950b1e6ffcaeb82a25e026f6cd581197debacc075c53fcbc250018bf073199b", "generation_seconds_batch": 18.051342872087844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5293986797332764, "target_z": -0.9570312583942484, "ce": 6.260893821716309, "genuineness": 4.482243061065674, "identity": 0.5628025531768799, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/557def5078ebb2c18c25c9f74b9a9c6b1f3dd6736c504840dba1ce90ef4b58a6.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "55a2d3dc35dfff15b32292e23620319434b41b1ec1a48f95862ce1152c071519", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aa0d112262f2162ffef175be32f37e4f4eed08a16ddc73c6ae97230db0a0287c", "generation_seconds_batch": 16.1072015799582, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.813485622406006, "target_z": -0.7343748905098032, "ce": 5.780362129211426, "genuineness": 4.532451152801514, "identity": 0.5527284145355225, "hypothesis": "Ich hab die kleine Schachtel gefunden und sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.125, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/55a2d3dc35dfff15b32292e23620319434b41b1ec1a48f95862ce1152c071519.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": "5655d9af662991ae5ae2da693128749eecf1679d38f24dd5210f71ac691f133c", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bd25e137b99b77721fa68555af7540e1072a7feefb21e0e1341b81c924f6d335", "generation_seconds_batch": 17.161306614987552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7629334926605225, "target_z": 0.009765661314248613, "ce": 5.629451274871826, "genuineness": 4.595211505889893, "identity": 0.6345494389533997, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/5655d9af662991ae5ae2da693128749eecf1679d38f24dd5210f71ac691f133c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "565c8529d28d7c999938e54cf06d1e376e761cd9546da2716e1835a420d371e6", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "f220b7d0bfd51c37b91adb8932a25743ea2dd4d75fc88a502afb8c83cf1857e5", "generation_seconds_batch": 22.66537852305919, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.28696346282959, "target_z": -0.36328125547450985, "ce": 6.006728172302246, "genuineness": 4.432034492492676, "identity": 0.7616869807243347, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/565c8529d28d7c999938e54cf06d1e376e761cd9546da2716e1835a420d371e6.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "567bd451b27d5601d99229322c05bf479d022f3e5d3bfd1c1940047f51ef6a91", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "fa06d2db61d5c0c0eb1758d941f0278fa8c433ef42ae6cc8aa0848044979f739", "generation_seconds_batch": 22.850207146955654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6938700675964355, "target_z": -0.8281249401453591, "ce": 6.149696350097656, "genuineness": 4.444586753845215, "identity": 0.830062210559845, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/567bd451b27d5601d99229322c05bf479d022f3e5d3bfd1c1940047f51ef6a91.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "568c5a9cab6a59f1af4773f1fb9fa49097e57db4fc7dc5f103f616ea7685cdf4", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2269f5dcbdaa3de241af5bd44deda4b7be6e21fc436639d740f3413fef677703", "generation_seconds_batch": 23.25060064601712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.813485622406006, "target_z": -0.7343748905098032, "ce": 6.173524379730225, "genuineness": 4.262581825256348, "identity": 0.25988879799842834, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/568c5a9cab6a59f1af4773f1fb9fa49097e57db4fc7dc5f103f616ea7685cdf4.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "56ff55647440cd3ba199f54a439445a2f79cc2cccea17d2b073c9b944705b820", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "a795df32f55d7e06c22f600fd4e61596b98cd540771b771dcdf2972ab86a89da", "generation_seconds_batch": 16.297755577019416, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.305119514465332, "target_z": -1.1328126014609157, "ce": 6.165581703186035, "genuineness": 4.306513786315918, "identity": 0.5812082886695862, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/56ff55647440cd3ba199f54a439445a2f79cc2cccea17d2b073c9b944705b820.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "5728f1d11365ea13aba9aefb6e745fd5f9f82194340070ed395c1728cdaf116b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "83445c5c08ff6e36e817e9ce8a8f620286f1c7556063657af3438b9c83f4f325", "generation_seconds_batch": 22.234044675016776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0028767585754395, "target_z": -0.5859374364956859, "ce": 5.959072113037109, "genuineness": 4.6705241203308105, "identity": 0.44473206996917725, "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_CASU_reg/5728f1d11365ea13aba9aefb6e745fd5f9f82194340070ed395c1728cdaf116b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "572ea0608b13630e15282876a417fd3390b0dad136bd80c7b76878015c3aae9e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "e74b6a3a285e8dcd5109069dd195d489917fa63e08f8384f3790ccb9876776e0", "generation_seconds_batch": 23.410271295928396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8433892726898193, "target_z": -0.7109375649641835, "ce": 6.149696350097656, "genuineness": 4.20609712600708, "identity": 0.2142219990491867, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/572ea0608b13630e15282876a417fd3390b0dad136bd80c7b76878015c3aae9e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "57476e32b95aad68bc8d91f1c98561b19656735308d18f7bef084491263c775e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "19c0e2d33d8bf2a3d8f40130548cf40017dbf4f979408c5c9add4463a0100b66", "generation_seconds_batch": 27.112057235906832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.022100448608398, "target_z": 0.21289070657019663, "ce": 6.109982967376709, "genuineness": 4.7332844734191895, "identity": 0.6063274145126343, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/57476e32b95aad68bc8d91f1c98561b19656735308d18f7bef084491263c775e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "57619de7e63c72d1c33857312991ed891f20611cb0fef8ddf05276e2bfdc417f", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b470c8b7c64d28913b142a016332cbbf229ca9c2e4e16438519f51b85d6be23e", "generation_seconds_batch": 22.053056915057823, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5279741287231445, "target_z": 0.6093749489045748, "ce": 5.748591423034668, "genuineness": 4.494794845581055, "identity": 0.21709269285202026, "hypothesis": "Ik heb het klein doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/57619de7e63c72d1c33857312991ed891f20611cb0fef8ddf05276e2bfdc417f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "57fa0b9f3be739b31b46cd16f5a368b607685d27a30fa03903b1cbb36d6357d5", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "fe12d40aed18e1e33a3d0ea8e0a75c71922085bf80e0bdbb2b45a1f4f8b62113", "generation_seconds_batch": 22.949997978983447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.264535665512085, "target_z": -0.38085929634954196, "ce": 6.125868320465088, "genuineness": 4.1998209953308105, "identity": 0.21335472166538239, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/57fa0b9f3be739b31b46cd16f5a368b607685d27a30fa03903b1cbb36d6357d5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "585e461b24f6fe4e005b0cd710125c6852cecf5fe216323edeffee890c19ce48", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "133bb8a17556a708613f6b40a96df24a4382aaea80a57dd46511e5e7025f55cb", "generation_seconds_batch": 21.701490371953696, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7324066162109375, "target_z": -0.014160124817189328, "ce": 5.95112943649292, "genuineness": 4.582659721374512, "identity": 0.5404605865478516, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/585e461b24f6fe4e005b0cd710125c6852cecf5fe216323edeffee890c19ce48.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "5873224814950c9451bc28ea707b176bd2c4921072b246e6e1c78a4801e0f889", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "d2f43b276c934cb773a3e2e3cabfb466b0521024b872536d390cccf06aa7e344", "generation_seconds_batch": 18.040603850968182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.577814102172852, "target_z": 0.6484375941615693, "ce": 6.1020402908325195, "genuineness": 4.482243061065674, "identity": 0.6767829656600952, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/5873224814950c9451bc28ea707b176bd2c4921072b246e6e1c78a4801e0f889.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "58acabb02c98806d53c86f61420fa4edf20916e1d0f423df92464a931545170b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "34515408e922a5529462ad16532121addef23ebc5c474d4b8491c37585543484", "generation_seconds_batch": 22.906182117993012, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.268807411193848, "target_z": 0.40625009051189603, "ce": 6.165581703186035, "genuineness": 4.519899368286133, "identity": 0.48840397596359253, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/58acabb02c98806d53c86f61420fa4edf20916e1d0f423df92464a931545170b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "58df1a22d3236db356b62c595494309bbea3130f2f102f87694fe0b182d7238a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "665eeff9ed3a9abca7bdd64d243bddeceb649b833933e72498b396768c4eeb4f", "generation_seconds_batch": 24.396074668038636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8439230918884277, "target_z": 0.07324217956196073, "ce": 6.24500846862793, "genuineness": 4.519899368286133, "identity": 0.7105309963226318, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/58df1a22d3236db356b62c595494309bbea3130f2f102f87694fe0b182d7238a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "5900b6ff1984b389c25c02e4a8d4a1687b4bfc74458c9dc2db6ec73446ae70f6", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "2500a671e48327055d35f3f5785f0af6a4eb7000972fe0b5ed84e97273e0bed9", "generation_seconds_batch": 23.810764549998567, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.3410749435424805, "target_z": 0.46289058978065334, "ce": 6.054384231567383, "genuineness": 4.545003414154053, "identity": 0.4575650990009308, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/5900b6ff1984b389c25c02e4a8d4a1687b4bfc74458c9dc2db6ec73446ae70f6.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "59582a533e5ef90760ca3ceb6f4716aea9cac0314fd0f4b1915d1dbac60cff72", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "e611a70b1bea8f337319e18d65652d69bd227273e545b8fb90e79a3f4416bd29", "generation_seconds_batch": 22.499581065028906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.079416275024414, "target_z": 0.2578126364977787, "ce": 5.681078910827637, "genuineness": 4.356721878051758, "identity": 0.16194027662277222, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/59582a533e5ef90760ca3ceb6f4716aea9cac0314fd0f4b1915d1dbac60cff72.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "5958c5df584447ec14d4d924eff04deb32c78da91e1a3c7915e43937bf23a396", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "9d5c874d615635a2702b363dcce1ddc8bf94cda3c1ce16df3fddc5e50f1eb4fc", "generation_seconds_batch": 24.424999363021925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.089384078979492, "target_z": 0.2656250160585622, "ce": 6.221180438995361, "genuineness": 4.645420074462891, "identity": 0.6757523417472839, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/5958c5df584447ec14d4d924eff04deb32c78da91e1a3c7915e43937bf23a396.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "597b9580fdfab9f2e190406100619130f84953a9109e200d1c2f33b940f1fcce", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4620158b91e65310e2ff7ece6b3b731c3d03f6f465297546a412bc951123527d", "generation_seconds_batch": 23.46913131303154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3250555992126465, "target_z": -1.1171874686128103, "ce": 6.213237762451172, "genuineness": 4.595211505889893, "identity": 0.6015370488166809, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/597b9580fdfab9f2e190406100619130f84953a9109e200d1c2f33b940f1fcce.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": "59d87cc5cb24c0d0822d727d4a35753b3f0119b2f8a20b80052ce4399852da35", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6da86bc7ec400952fd79f068e3d78b81f798b245cc2c45d1ccbc72fd5e14f9df", "generation_seconds_batch": 17.656199467019178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.592855215072632, "target_z": -0.12353512043758146, "ce": 6.078212261199951, "genuineness": 4.758388519287109, "identity": 0.5344812273979187, "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_CASU_reg/59d87cc5cb24c0d0822d727d4a35753b3f0119b2f8a20b80052ce4399852da35.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "5a52cadb13e983f6547a7c5388411c286111e55b42e438db334049946e353696", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "07a39eec23a45080eaae095c9225bc175887fb9c0bb66c3c33cfea92f0da2679", "generation_seconds_batch": 16.433814181014895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.364927291870117, "target_z": -1.0859375766431378, "ce": 6.38003396987915, "genuineness": 4.68307638168335, "identity": 0.6132502555847168, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/5a52cadb13e983f6547a7c5388411c286111e55b42e438db334049946e353696.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": "5ac62e8d4c1c88b8d5435f2e15ee7950d9ae549e17f3fe918d51e2f7176e1d98", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ee5b30103f19df5fbdce29d08ee7b6109c551e885f2a819731cecc59663e54bc", "generation_seconds_batch": 22.96834774001036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0046563148498535, "target_z": 0.19921866861228701, "ce": 5.982900142669678, "genuineness": 4.6203155517578125, "identity": 0.6380851864814758, "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_CASU_reg/5ac62e8d4c1c88b8d5435f2e15ee7950d9ae549e17f3fe918d51e2f7176e1d98.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "5ad177109ecc5ea937dbb88bf0684e1ffd7b651bc79eafb485e3ddf3505a6997", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "0c8b7f542840d205f637898f61a3f2d2b437bb7f497cb80487508bbd8cc14f89", "generation_seconds_batch": 23.715953538077883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.17909574508667, "target_z": 0.3359375532852291, "ce": 6.3641486167907715, "genuineness": 4.6203155517578125, "identity": 0.4275784492492676, "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_CASU_reg/5ad177109ecc5ea937dbb88bf0684e1ffd7b651bc79eafb485e3ddf3505a6997.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "5b8473cc9e9cf94d76f61ef4931c125c32a4fc1470b7297f55b5c844601ddd69", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0542d3b24bc298027ba65549d13598421648f3d41860bc0cc9cdac0eaf67492b", "generation_seconds_batch": 22.299424338969402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.281267166137695, "target_z": 0.4160155649628754, "ce": 5.621508598327637, "genuineness": 4.582659721374512, "identity": 0.7541229724884033, "hypothesis": "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_CASU_reg/5b8473cc9e9cf94d76f61ef4931c125c32a4fc1470b7297f55b5c844601ddd69.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "5bbc8d4d8abe5fae425ca69c8ad4d96ea05e2b4cd335a21d7938659621c782d2", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "34633de65530d521983fe78ce12504ea0e8f580d7ea95c911db09f4a979c8f72", "generation_seconds_batch": 21.70309393608477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.95855450630188, "target_z": 0.16308585255385563, "ce": 5.871703147888184, "genuineness": 4.595211505889893, "identity": 0.3591828942298889, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/5bbc8d4d8abe5fae425ca69c8ad4d96ea05e2b4cd335a21d7938659621c782d2.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "5bce1ff60794536b4ce2c1e76270d5ffc4f2e544a8bfe5a45863a81584d0f589", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "51bd7eb2a399a97d2ff87dd88cdae55f726b48f7cd9afdf15b45d133297b6854", "generation_seconds_batch": 22.361798034980893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.924912691116333, "target_z": 0.13671869780967286, "ce": 6.086154937744141, "genuineness": 4.482243061065674, "identity": 0.7335443496704102, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/5bce1ff60794536b4ce2c1e76270d5ffc4f2e544a8bfe5a45863a81584d0f589.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "5c080bb6b9b87e702f767726022357c4786909f3cd35417a3edd78c798012eae", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "18008d9841ca7d524beca561adddb2638fb8bd09e36f1990a3331db1b4c54b0e", "generation_seconds_batch": 23.96699068392627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9562411308288574, "target_z": -1.4062499970802613, "ce": 6.252951145172119, "genuineness": 4.20609712600708, "identity": 0.2894098162651062, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/5c080bb6b9b87e702f767726022357c4786909f3cd35417a3edd78c798012eae.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": "5c171a1e37120cc463344ef66d4a3e14b138848b1ebe1fc30c921e7ae4d0c6b6", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4c82aa024f00f0b8d2174a008b25624e3ce0e499a0df86843d2ba122f7033f95", "generation_seconds_batch": 17.667625197093002, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.58778190612793, "target_z": 0.6562499737223527, "ce": 5.990842819213867, "genuineness": 4.469690799713135, "identity": 0.6458747982978821, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/5c171a1e37120cc463344ef66d4a3e14b138848b1ebe1fc30c921e7ae4d0c6b6.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": "5c1865f5160685c625ddfd0c783609db3d6a3c58918533b263dc122798fa669a", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Escribí el número en el sobre.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f324ba1af2db3ff7a127d91036e58b842c20ddbad799915282628b9f59a80ff", "generation_seconds_batch": 24.85774082201533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.288743019104004, "target_z": 0.421874849633463, "ce": 5.839932441711426, "genuineness": 4.7081804275512695, "identity": 0.6156832575798035, "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_CASU_reg/5c1865f5160685c625ddfd0c783609db3d6a3c58918533b263dc122798fa669a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "5c30b25bad3549b2852bf85e0cee4fb1c864b5d62f53cad6ade8cb60da566a5f", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "8266e892105a5c89661fb5662912748f1ff88bd33b46a74f300b005db3a2c375", "generation_seconds_batch": 17.84299272496719, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.169127941131592, "target_z": 0.3281251737244456, "ce": 6.221180438995361, "genuineness": 4.482243061065674, "identity": 0.5865888595581055, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen daar samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/5c30b25bad3549b2852bf85e0cee4fb1c864b5d62f53cad6ade8cb60da566a5f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "5c44b478415e3836275ff4463bf32a2e788a1417c452b303c778ba0d47fb5318", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "4ce9fac45db553b8fc7661418d4302b4052b0a3bcb813e9c1d70e20e3d415aeb", "generation_seconds_batch": 23.588325946009718, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.772278308868408, "target_z": 0.017089767152483136, "ce": 5.605623245239258, "genuineness": 4.444586753845215, "identity": 0.7983801364898682, "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_CASU_reg/5c44b478415e3836275ff4463bf32a2e788a1417c452b303c778ba0d47fb5318.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "5c52f232f8d0e148fec2365a17a163b4c44bc435f8fa6bf63ddaa02268e47c8c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d145e22d6fdbc1f91d64ab03fc2a6cd7c322a15b7118d1c9256683ae40dd62e", "generation_seconds_batch": 24.49523817107547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0758566856384277, "target_z": -1.3124999474447054, "ce": 5.7684478759765625, "genuineness": 4.20609712600708, "identity": 0.3781351149082184, "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_CASU_reg/5c52f232f8d0e148fec2365a17a163b4c44bc435f8fa6bf63ddaa02268e47c8c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "5cd558e09cb7d9d69a322ac33ce673b69cdaac7fbf3aa146d70754bcec4fb5cd", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "cc865e4eb992dd2b05579b8bb07a29fc0be006303ea64abea4e03953671bf4e7", "generation_seconds_batch": 16.05221094901208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4663867950439453, "target_z": -0.222656181021176, "ce": 5.903473377227783, "genuineness": 4.293961524963379, "identity": 0.6852903962135315, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/5cd558e09cb7d9d69a322ac33ce673b69cdaac7fbf3aa146d70754bcec4fb5cd.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "5d3573e493e479a53ffeaf4e2498602bf476e7d11d88002ba40fc5a1d4c749ed", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2070d735b9334f81a95372bd9d58880348a8e59c6ba1c0d4e1059fdbfb750dc7", "generation_seconds_batch": 22.664694448933005, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.972972869873047, "target_z": -0.6093749489045748, "ce": 5.831989288330078, "genuineness": 4.005263805389404, "identity": 0.42795801162719727, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/5d3573e493e479a53ffeaf4e2498602bf476e7d11d88002ba40fc5a1d4c749ed.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "5dd08f35b71f641933dd00e329798e3f10cae71713cf945dea984413d3ea0fdf", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "c4f1f7f85fa66df379e067d7f867bb98085d20d09ac0971f5f1c4b3b51fd6e5f", "generation_seconds_batch": 22.014224562095478, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9480528831481934, "target_z": -0.628906271533072, "ce": 6.133810997009277, "genuineness": 4.369274139404297, "identity": 0.5725593566894531, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het Oude Raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/5dd08f35b71f641933dd00e329798e3f10cae71713cf945dea984413d3ea0fdf.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "5b943216f4a1cc3e99086eec22811fe7df64c7aab06fbf953f29bd6a8b26e5ba", "generation_seconds_batch": 17.592919258982874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9929087162017822, "target_z": -0.5937500029197386, "ce": 6.109982967376709, "genuineness": 4.482243061065674, "identity": 0.8420531749725342, "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_CASU_reg/5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "5fa696918ea88366ee9260abf213fccfd4b432b2900acbf7943d2e59ec778437", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c351686fb18b59bd460cb9c39244b5aba879e5b619e847d0aa54987740d4efd5", "generation_seconds_batch": 23.177522643003613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.524414539337158, "target_z": -0.9609376350379094, "ce": 6.09409761428833, "genuineness": 4.545003414154053, "identity": 0.6707905530929565, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/5fa696918ea88366ee9260abf213fccfd4b432b2900acbf7943d2e59ec778437.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "5fb95fd310eca570fed66a6da69044b0497aaf3d9873eb06e19741b1a3f89192", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "83a4960efe1270cd5e37a77f64efff8c7511c332de722fc001948374ba8ba121", "generation_seconds_batch": 22.312673361971974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2652478218078613, "target_z": -1.164062493430588, "ce": 6.260893821716309, "genuineness": 4.281409740447998, "identity": 0.5002244710922241, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/5fb95fd310eca570fed66a6da69044b0497aaf3d9873eb06e19741b1a3f89192.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "6071a901c36fe35d57f19ef1475095d93bcff72e4841270ea499ebdc29085204", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d9148bb185758d11373b5d664acfec81d1105b503d1ea8ead438e77987a0bdef", "generation_seconds_batch": 16.90145147300791, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.846948623657227, "target_z": 0.8593748321150315, "ce": 6.070269584655762, "genuineness": 4.432034492492676, "identity": 0.6819908022880554, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/6071a901c36fe35d57f19ef1475095d93bcff72e4841270ea499ebdc29085204.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "60a4f9479d5664f423f21df8064433a8b9c9da89d2e94cce70af7e27d6628ac0", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1022d5d63ea2eb1d5208035ee01f24a1cba3afba4567d3bfcb50af562e34b58c", "generation_seconds_batch": 22.687444000970572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.588494300842285, "target_z": -0.12695303649542422, "ce": 5.125091552734375, "genuineness": 4.469690799713135, "identity": 0.7569446563720703, "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_CASU_reg/60a4f9479d5664f423f21df8064433a8b9c9da89d2e94cce70af7e27d6628ac0.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "60aec4b8f911cae5832349314d703347b8a65b3678c06f029a90b68f938e3c17", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "2ea1835e53106d51edf2c1a4e8196b87eb11d57c44c0bb5f9447daedfc1ff4fb", "generation_seconds_batch": 17.40407596004661, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.247091770172119, "target_z": -0.39453114744418233, "ce": 5.998785495758057, "genuineness": 4.595211505889893, "identity": 0.5296115875244141, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist,", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/60aec4b8f911cae5832349314d703347b8a65b3678c06f029a90b68f938e3c17.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "60b188889002b7d653e1d4487973cb346d1a107e9309e053256b11da9178eeac", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "58a1e3331a84345ba8a5fa9515bbec2e75a0e0d1fc8d983adf407c9865f5f59b", "generation_seconds_batch": 21.429744058987126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.729447364807129, "target_z": -0.016479471715114238, "ce": 5.7247633934021, "genuineness": 4.758388519287109, "identity": 0.6443575024604797, "hypothesis": "Non, c'est. J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/60b188889002b7d653e1d4487973cb346d1a107e9309e053256b11da9178eeac.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": "610d933790a514ac057b38ef4b7fc10baf321f8a50ed27ecff0a6fe8f54b332c", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e026b098a2e00d91e8cab0c13f9f1ed1e6dff087a989c7723c0865773ad38c26", "generation_seconds_batch": 16.67027512995992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.139224052429199, "target_z": 0.30468766131555663, "ce": 6.062326908111572, "genuineness": 4.545003414154053, "identity": 0.6937484741210938, "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_CASU_reg/610d933790a514ac057b38ef4b7fc10baf321f8a50ed27ecff0a6fe8f54b332c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "6136f7e87a946e5ca288b2e9929137d8dce2964845f2934ca296c4523343248e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9aad882e8d324411ce0f461dac342a70457aa5e77cc2eb3a155b18b9c932ea80", "generation_seconds_batch": 26.342725471011363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.36919903755188, "target_z": -0.2988281897816998, "ce": 5.990842819213867, "genuineness": 4.632867813110352, "identity": 0.8192662596702576, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/6136f7e87a946e5ca288b2e9929137d8dce2964845f2934ca296c4523343248e.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": "61ab00cb792445e60a34f393874d3a0dacaf2026a9adba935923801551f91378", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e43b80acde61b794d5ae3bbcb2b6adf257f14a64f1f0bbba795e840190fe5043", "generation_seconds_batch": 16.664834236027673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.291947364807129, "target_z": -0.3593750656941181, "ce": 6.086154937744141, "genuineness": 4.7081804275512695, "identity": 0.614967942237854, "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_CASU_reg/61ab00cb792445e60a34f393874d3a0dacaf2026a9adba935923801551f91378.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "6278ef179bb7ab3b4237c1fd2315d9d7a5974ce53e2b7894f45d7e6be4893dac", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4adcb604d5d88ec810d5659247e4d1eec542747ba3cc7245c9cb8b8f484bd4ca", "generation_seconds_batch": 22.59607640898321, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7967537641525269, "target_z": -1.5312500321171243, "ce": 6.403861999511719, "genuineness": 4.3504462242126465, "identity": 0.5159782767295837, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/6278ef179bb7ab3b4237c1fd2315d9d7a5974ce53e2b7894f45d7e6be4893dac.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "62b9802ac1065b76a7538ca9664bafec324ffc980f7e15d3d2e3af1da294f9a6", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c10ff965fc6d01d672f6ae200c31661a9acf16c037fff1b5a897e8e249a0155", "generation_seconds_batch": 22.433138379012235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5229902267456055, "target_z": 0.6054687591241831, "ce": 5.839932441711426, "genuineness": 4.469690799713135, "identity": 0.5419936180114746, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/62b9802ac1065b76a7538ca9664bafec324ffc980f7e15d3d2e3af1da294f9a6.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "6310e1bb39c754d38f3784466bc46809d11aef7f2edc9069a10f8a7146f49fcc", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "69b190c374a52e5de029332793a6a09c33f15bf3d6f87627b8f49202c06f3b87", "generation_seconds_batch": 21.745204703998752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5443506240844727, "target_z": -0.9453125021898039, "ce": 5.796247482299805, "genuineness": 4.331617832183838, "identity": 0.7634195685386658, "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_CASU_reg/6310e1bb39c754d38f3784466bc46809d11aef7f2edc9069a10f8a7146f49fcc.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "63810edc10e672f8e5c81b6558b474e1c6f588e216e329a06108820b0eb0a50f", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "1c05e69e170ff5eb8c59491ccf4993b4ef0700369038d6e86cf1a23d3d544b01", "generation_seconds_batch": 22.008470964967273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6073625087738037, "target_z": -1.6796875795628763, "ce": 6.38797664642334, "genuineness": 4.030367851257324, "identity": 0.5468553304672241, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/63810edc10e672f8e5c81b6558b474e1c6f588e216e329a06108820b0eb0a50f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "63c38756eb68f1c29f8b5b018c9f0b0b44d5249791bb315f5c9eb2321ad44e12", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c9b94a789d2d416a8a0b8ff117df35bf3235cb61cf5db5b30ede352759b76480", "generation_seconds_batch": 23.682568341959268, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6988539695739746, "target_z": -0.8242187503649673, "ce": 6.141753673553467, "genuineness": 4.457139015197754, "identity": 0.6982169151306152, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/63c38756eb68f1c29f8b5b018c9f0b0b44d5249791bb315f5c9eb2321ad44e12.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "63d20f923876fc407f06a8a6b760baa0156f4d85f64fbdff0e5a61263bb18a1e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "0fa735dacb080eccdefdad8ad2e840e4bbf470ec3ec0d67c02ab1c086b827f68", "generation_seconds_batch": 18.006599662010558, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.065710067749023, "target_z": 0.2470702408751629, "ce": 6.268836498260498, "genuineness": 4.394378662109375, "identity": 0.39802801609039307, "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_CASU_reg/63d20f923876fc407f06a8a6b760baa0156f4d85f64fbdff0e5a61263bb18a1e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "64342431794045a1e478ee52720b034faa97bb1152aa7fda5646d17befb1f126", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "744ea181d543a2b1e70ec4e6acc3e0fad87bab56af1e0e048d48df5393705b4c", "generation_seconds_batch": 21.620909058023244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8533573150634766, "target_z": -0.7031249985401307, "ce": 6.221180438995361, "genuineness": 4.425758361816406, "identity": 0.3658929467201233, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/64342431794045a1e478ee52720b034faa97bb1152aa7fda5646d17befb1f126.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "644227d62a5813429915b2fb971c2b382eaa3b72f91cbea282e5591b36d70097", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "64937ad68e066b76643c7dde52516130617bbf1bd68858e8f79fa240b5e90272", "generation_seconds_batch": 22.242438716930337, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.012132167816162, "target_z": 0.20507795328287465, "ce": 5.700935363769531, "genuineness": 4.883909225463867, "identity": 0.8088201880455017, "hypothesis": "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_CASU_reg/644227d62a5813429915b2fb971c2b382eaa3b72f91cbea282e5591b36d70097.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "64c2c14949eeaa47e556d387573a1825cad5130fb6dbc1b9be5d490285dcc316", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4e6368a303e942103012552afdbd050ebecf3d0175035d0baf3deb7247ce1074", "generation_seconds_batch": 23.29418758803513, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.759195566177368, "target_z": 0.0068360189789548045, "ce": 6.022613525390625, "genuineness": 4.432034492492676, "identity": 0.7210015058517456, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/64c2c14949eeaa47e556d387573a1825cad5130fb6dbc1b9be5d490285dcc316.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "65895e34a66ceea60314bc6221808e76fedf36723f72ddf6e8ddebb467312575", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7818f3ff646ce4b83ed244796079b52b2dfa3658d1f11cfa45c6a7b2d40cf70b", "generation_seconds_batch": 16.657624796964228, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.092588424682617, "target_z": -0.5156248992690189, "ce": 6.0146708488464355, "genuineness": 4.400654315948486, "identity": 0.6923686265945435, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/65895e34a66ceea60314bc6221808e76fedf36723f72ddf6e8ddebb467312575.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "66779842fd728d0b699d6e660a74c4b697d20616f14bd9ebab366ef226f05938", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "3f4dd354e261394a842ecc253c27f4cc2e2948eecbae62edda38887cae0c170b", "generation_seconds_batch": 22.192778203985654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.050758361816406, "target_z": 0.23535167153398767, "ce": 6.062326908111572, "genuineness": 4.394378662109375, "identity": 0.5060880184173584, "hypothesis": "Ik heb het klein nedoosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/66779842fd728d0b699d6e660a74c4b697d20616f14bd9ebab366ef226f05938.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "66913a0da5a25543cf38e1e87d108823576234d818d23abb9a8d98cd10cdc3d5", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a507ae8385baaa790ba12cf65e0edf2bf216fb3d9d962423abb619094afcd2a7", "generation_seconds_batch": 22.227656582952477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.109320163726807, "target_z": 0.28125014890666766, "ce": 5.967014789581299, "genuineness": 4.6705241203308105, "identity": 0.6778589487075806, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/66913a0da5a25543cf38e1e87d108823576234d818d23abb9a8d98cd10cdc3d5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "66d87f3d8cd12b66ac5e180a9ccc9a27de51bd4aaeea424cda0933c7633bf47d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "dfee70fd4997ef84d37081e34d0a82aa44cc22b9c37a4f6c9fba48cec4757321", "generation_seconds_batch": 23.047014658921398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.195472240447998, "target_z": -1.2187498978091496, "ce": 6.23706579208374, "genuineness": 4.406930446624756, "identity": 0.4800386130809784, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/66d87f3d8cd12b66ac5e180a9ccc9a27de51bd4aaeea424cda0933c7633bf47d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "6739b99f0cc00fb840e3b21d3ad23365ee144b21e1261871d54be13b1815d05b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "99b6f3a20856984cba770409e094a77f09a5157dbae13f516e9a4efebecf0ce5", "generation_seconds_batch": 16.646278284024447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8482840061187744, "target_z": 0.07666009561980351, "ce": 6.149696350097656, "genuineness": 4.432034492492676, "identity": 0.5016807317733765, "hypothesis": "Ik 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_CASU_reg/6739b99f0cc00fb840e3b21d3ad23365ee144b21e1261871d54be13b1815d05b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "67a6a79de52ce9d894bd4c7d7ce80049a202c1be019a2145c9c2ddc8bb5dce40", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "91aaf241dee9e7b1bae553405613b7ab62009133d8d8f08da811d16977aca87d", "generation_seconds_batch": 22.556334809982218, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0758566856384277, "target_z": -1.3124999474447054, "ce": 6.24500846862793, "genuineness": 4.432034492492676, "identity": 0.5889145135879517, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/67a6a79de52ce9d894bd4c7d7ce80049a202c1be019a2145c9c2ddc8bb5dce40.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "67c4b895898393c50486e9fd42b2ea26a9f1090e090b14aa8095d79cd413cac4", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "bf26b1cbae25804b474dcbca3ed8c4e9972150d198abb150ec4d06df304a1446", "generation_seconds_batch": 22.620094978949055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9263372421264648, "target_z": -1.4296875094891504, "ce": 5.736677646636963, "genuineness": 3.923675537109375, "identity": 0.5024811029434204, "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_CASU_reg/67c4b895898393c50486e9fd42b2ea26a9f1090e090b14aa8095d79cd413cac4.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "67dbd818dc6efc41a116b691e2cafebd1033a979d887a8a91f363e4ffd44072c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "452166370950ac51565919aac1149cd985a4ff052801b4fa521215c89198ea16", "generation_seconds_batch": 17.290432508918457, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4713706970214844, "target_z": -0.21874999124078426, "ce": 5.839932441711426, "genuineness": 4.7458367347717285, "identity": 0.6516213417053223, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/67dbd818dc6efc41a116b691e2cafebd1033a979d887a8a91f363e4ffd44072c.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": "67e701b73aac39d7edd5bfe4909722459e704fa7af0a71a34844b2ab27683b47", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "be2129d836896f92d9346ad196477b9cd57fcf4d8203a7adc8afa88ddc257194", "generation_seconds_batch": 18.37050955905579, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2720115184783936, "target_z": -0.3750000116789543, "ce": 5.804190158843994, "genuineness": 4.362998008728027, "identity": 0.5392276048660278, "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_CASU_reg/67e701b73aac39d7edd5bfe4909722459e704fa7af0a71a34844b2ab27683b47.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "68886592a43adf4b32c82d2cbcaf2db73e6fbcd8d412efcda857359486b66f4f", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "1ad20d3ac7f97994202901e3d744f80ed00d53f55e16d2d825b171b486875489", "generation_seconds_batch": 22.687641764990985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5941903591156006, "target_z": -0.9062500437960788, "ce": 6.205295085906982, "genuineness": 4.074300289154053, "identity": 0.3598831295967102, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/68886592a43adf4b32c82d2cbcaf2db73e6fbcd8d412efcda857359486b66f4f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "68d32b770e01383c9cba1f0d98d5201a449a15c928ecf9157adf80ce91bebc8c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "e6b92fa83ad22d32d7bb67ca4e79893121b5e9967274be6d2f6127c3204e5e65", "generation_seconds_batch": 24.55935576907359, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.511242389678955, "target_z": -0.18750009927111178, "ce": 5.959072113037109, "genuineness": 4.381826400756836, "identity": 0.7521531581878662, "hypothesis": "He encontrado la cajita. estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/68d32b770e01383c9cba1f0d98d5201a449a15c928ecf9157adf80ce91bebc8c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "693b14ece1e97d80770ed3fa6dbd6126df782d6b2e62af03474346183467188d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "79c6bbabd29197afaa1f6bef95b8ba0e020520a04da9ddc387eaee46895a251a", "generation_seconds_batch": 16.738397576031275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.483118534088135, "target_z": 0.5742188671545106, "ce": 5.665193557739258, "genuineness": 4.595211505889893, "identity": 0.7498663663864136, "hypothesis": "La puerta sigue abierta. he dejado tu abrigo, es sobre la silla. Dime cuándo estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/693b14ece1e97d80770ed3fa6dbd6126df782d6b2e62af03474346183467188d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "69fdcece3ee49bc9ba539bd689197199f2031fd56b35c15eecde095bf8815610", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "770eaf9b4bc3da33ebabfe168932d15f906ddbe3bba5caac579b68f828c51002", "generation_seconds_batch": 23.575429996009916, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.454638957977295, "target_z": -1.0156250394164708, "ce": 6.348263263702393, "genuineness": 4.381826400756836, "identity": 0.6396174430847168, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/69fdcece3ee49bc9ba539bd689197199f2031fd56b35c15eecde095bf8815610.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "6a1ee5d531c687f35c22e78ae44fea4c172c7f848092e37b5a4606ce161b4cc7", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ff23f598449610bd4896c4d75f8bc03d7d843d7d46f526a2ba5f9cb8e81cc5a6", "generation_seconds_batch": 21.913127915002406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5137345790863037, "target_z": -0.18554681751764668, "ce": 5.967014789581299, "genuineness": 4.105680465698242, "identity": 0.43807733058929443, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/6a1ee5d531c687f35c22e78ae44fea4c172c7f848092e37b5a4606ce161b4cc7.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "6a40aa515af2c9b6c792d0bda5f31817e3eb272242cdf2efa074c19ff882ed3b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "18e003d8a63c1983c8abf297e2991999d4553cc55b094ae2b54852a8ae093525", "generation_seconds_batch": 22.766464699991047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.702502727508545, "target_z": -0.0375976372260783, "ce": 5.863759994506836, "genuineness": 4.457139015197754, "identity": 0.47476449608802795, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/6a40aa515af2c9b6c792d0bda5f31817e3eb272242cdf2efa074c19ff882ed3b.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "f2553245d7c9b46aa9ab29595b045042068fb2b0033ac3a2eb637f56ee54aed7", "generation_seconds_batch": 16.83135937305633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1499040126800537, "target_z": -0.4707031562047061, "ce": 6.062326908111572, "genuineness": 4.042920112609863, "identity": 0.3510000705718994, "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_CASU_reg/6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "6a786203aa60ef2c5d478ef2e0f886d33a27763e6ec965991b42a5876313f41a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "24b443bc4ddf928a5166a804097e39647e85b7ebfea92ba5fc54c6d49a4b3a53", "generation_seconds_batch": 22.79403879190795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.444671154022217, "target_z": -1.0234374189772544, "ce": 6.23706579208374, "genuineness": 4.545003414154053, "identity": 0.5305927395820618, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/6a786203aa60ef2c5d478ef2e0f886d33a27763e6ec965991b42a5876313f41a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "6a9943c161f1971461d63c37f4c17b7d6fe9cb56b60f1b3671bff58bcc1e29d1", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "551f99a6fa377c9c13c00df1682244c90447faedd0ab7b53f8bb2194b9dab907", "generation_seconds_batch": 21.58761911804322, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.166635513305664, "target_z": 0.32617170510771126, "ce": 5.4944257736206055, "genuineness": 4.325342178344727, "identity": 0.19886448979377747, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/6a9943c161f1971461d63c37f4c17b7d6fe9cb56b60f1b3671bff58bcc1e29d1.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "6aa655b9e805fe60837b1219d0f8ce8c45477621cc07b44eba019127d0a1680d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "ddff1b05f040e9e39ebaf99e6d32a32a83ea4cd9efc563c99d9df051bd4c77ce", "generation_seconds_batch": 22.65880296996329, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.67197585105896, "target_z": -0.061523423357516246, "ce": 6.062326908111572, "genuineness": 4.256305694580078, "identity": 0.39592432975769043, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/6aa655b9e805fe60837b1219d0f8ce8c45477621cc07b44eba019127d0a1680d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "6ad4f3f466b2e54d055f8bbb25fbe2420dff5bd769f5a8a8621eaa5905d7f46c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "2bd964abfa13d961a7698d800ed2f84a2225b04af05238bf138c5b0b10cfb1c1", "generation_seconds_batch": 16.076712407986633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411562919616699, "target_z": -0.2656250160585622, "ce": 6.173524379730225, "genuineness": 4.8588056564331055, "identity": 0.6371340751647949, "hypothesis": "J'ai trouvé la petite boîte elle était près de de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/6ad4f3f466b2e54d055f8bbb25fbe2420dff5bd769f5a8a8621eaa5905d7f46c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "6b02eda886f92f5fc4ef5ccbee2e0e141410d4c523517b7480f24fc1b5d99a9c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a21cd5067ff9f97560a8fb777a5eca65b55ad3443bb41904afcc8452ebeed9f1", "generation_seconds_batch": 23.32972328993492, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.533670425415039, "target_z": -0.16992187153281044, "ce": 6.165581703186035, "genuineness": 4.375550270080566, "identity": 0.21962866187095642, "hypothesis": "The door is still open? I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/6b02eda886f92f5fc4ef5ccbee2e0e141410d4c523517b7480f24fc1b5d99a9c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "6b20139946bedd9ffc339ded75b5d311653136e09760b1034ff8b37966a635c7", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "f5fa384fa73908331405607e78d385481869324cde155e76a771142a360c77b2", "generation_seconds_batch": 24.828695817966945, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.281979560852051, "target_z": -0.3671874452549016, "ce": 5.967014789581299, "genuineness": 4.65797233581543, "identity": 0.6606345772743225, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/6b20139946bedd9ffc339ded75b5d311653136e09760b1034ff8b37966a635c7.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "cd2f55e147f0bbb2fa6a8504b6628f1a423372a41521f5d70e3f9505a3bdec63", "generation_seconds_batch": 16.36459890601691, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5149805545806885, "target_z": -0.18457027007254873, "ce": 6.205295085906982, "genuineness": 4.632867813110352, "identity": 0.6691436171531677, "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_CASU_reg/6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "6c49a8b1f1053a5e489293db3f3a70922745fc4cf5b81b9a5c93c9e19feec28a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "97a0b4a217670c1382732d074d3d1eeca3dfe5714ac323c3ea86976e5c7b7c1e", "generation_seconds_batch": 23.46319490496535, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.483830690383911, "target_z": -0.20898432992653565, "ce": 5.974957466125488, "genuineness": 4.545003414154053, "identity": 0.6229879260063171, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist?", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/6c49a8b1f1053a5e489293db3f3a70922745fc4cf5b81b9a5c93c9e19feec28a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "6e2b02ba69e4fe3e0744538b4bace2f8f6da58f9fe82bb1159194415df752835", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bc5a00091ba6f270fef52c42095e985d4397d441fd9bee674890f1bb1f9ad19b", "generation_seconds_batch": 23.377249875920825, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7593512535095215, "target_z": 0.006958040693774734, "ce": 5.824047088623047, "genuineness": 4.406930446624756, "identity": 0.747979998588562, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu y es prêt?", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/6e2b02ba69e4fe3e0744538b4bace2f8f6da58f9fe82bb1159194415df752835.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "6e565a300fb124cbb95e18cd920d4fa345962e73ad658495c8426bafad9bbaf6", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "1d2f08317ca71cac1d264b9225256715b67b24f8aa5f56523d2099e8312227fe", "generation_seconds_batch": 24.00280263996683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0128445625305176, "target_z": -0.5781250569349023, "ce": 5.990842819213867, "genuineness": 4.6956281661987305, "identity": 0.667839765548706, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/6e565a300fb124cbb95e18cd920d4fa345962e73ad658495c8426bafad9bbaf6.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "6e8ca08bf66e4544cffa4ef87dd0ada1ced0641476bfab0908de9888989d4f99", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "f7b408f7215468a8ed1b1db61d53b8981f49c2b1f6eefa7953e61822fa95fbcc", "generation_seconds_batch": 17.520284128026105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5635743141174316, "target_z": -0.14648435912392146, "ce": 5.355429172515869, "genuineness": 4.444586753845215, "identity": 0.7513934373855591, "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_CASU_reg/6e8ca08bf66e4544cffa4ef87dd0ada1ced0641476bfab0908de9888989d4f99.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "6ea7b61a141f61ee757584d6cfe415d5e968f660a313861a239e2cedd1c0899a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "cbee5b8050b1a362fc7186b5a16ee1ac3dd7b8eca03e95994f9eba5a6a99fed0", "generation_seconds_batch": 17.00756287202239, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.807076930999756, "target_z": 0.8281249401453591, "ce": 5.6254801750183105, "genuineness": 4.607763767242432, "identity": 0.20830278098583221, "hypothesis": "Ik heb een klein doosje gevonden. Het stond naast het oude ramen. We kunnen er samen in kijken.", "wer": 0.16666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/6ea7b61a141f61ee757584d6cfe415d5e968f660a313861a239e2cedd1c0899a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "6ea98bd431f92e104188c6684642543fcfed60a1b79266cde20efad0e11a0241", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "6fbb8b2bfb9f93e21e9e30e6fce257c44ca19f57761ed76e5e5fa8cb8dc121ed", "generation_seconds_batch": 21.547154724015854, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5536062717437744, "target_z": -0.1542969255479742, "ce": 5.839932441711426, "genuineness": 4.557555198669434, "identity": 0.472886860370636, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. 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_CASU_reg/6ea98bd431f92e104188c6684642543fcfed60a1b79266cde20efad0e11a0241.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": "6f51016d511d38d891ad927e972aa29ca7ce33dfff075d26d1bea3e36f26e327", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Someone moved the little blue chair.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ef2e4aa9b2e40ca96e142e1d500d85d0fe181c6b9d5659a74d5bd947976363b", "generation_seconds_batch": 17.950054474989884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.226443290710449, "target_z": 0.3730467299254892, "ce": 6.292664527893066, "genuineness": 4.6203155517578125, "identity": 0.3226354122161865, "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_CASU_reg/6f51016d511d38d891ad927e972aa29ca7ce33dfff075d26d1bea3e36f26e327.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "6f6242852300de312a0f1a35505bc5cebbeb4ee37b925600ef0e4c6181946f0a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "a5577b7a1aafa7b376619190ac1c18cf19ea01a135dd4a89258a17cf5bdf692d", "generation_seconds_batch": 25.03083440696355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.189063549041748, "target_z": 0.34374993284601263, "ce": 6.1020402908325195, "genuineness": 4.394378662109375, "identity": 0.7538313865661621, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/6f6242852300de312a0f1a35505bc5cebbeb4ee37b925600ef0e4c6181946f0a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "700fb8e4007cb26b9d58be2f145ba7ed12df26c06140cea4584b1f6df3bd93d6", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "10794ab428a5efbebf9daf9155b250f2d215ffb8167540bcae25200a315e4d91", "generation_seconds_batch": 25.061136021045968, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.658982276916504, "target_z": -0.8554686423346398, "ce": 6.062326908111572, "genuineness": 4.494794845581055, "identity": 0.3579365015029907, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/700fb8e4007cb26b9d58be2f145ba7ed12df26c06140cea4584b1f6df3bd93d6.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "702b9ba147d5b0c27e3ecaeebc41de2dd7385c28dda8a1a4104a9f150a1d3d8e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "982fa8c930ac7714587d8795f35a428747de9c281455634f5983b0932ca7c0ba", "generation_seconds_batch": 21.09516625793185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.823453426361084, "target_z": -0.7265625109490197, "ce": 5.935244083404541, "genuineness": 4.137060642242432, "identity": 0.3076459765434265, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/702b9ba147d5b0c27e3ecaeebc41de2dd7385c28dda8a1a4104a9f150a1d3d8e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "70712c64846ccbc0ac3e1d5e51e38f46c10a11afcecc511a18c7c2f8db2d0a9a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c7dd060345ed47b78745d0a1cb55b508804911e2fe16cc61f92cf64bc44d4b18", "generation_seconds_batch": 24.34107729594689, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8464150428771973, "target_z": 0.0751952744521566, "ce": 6.006728172302246, "genuineness": 4.808597087860107, "identity": 0.7684872150421143, "hypothesis": "Porte est encore ouverte j'ai posé ton manteau sur la chaise dis-moi quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/70712c64846ccbc0ac3e1d5e51e38f46c10a11afcecc511a18c7c2f8db2d0a9a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "70be3f105134a2516d1db5251a03d3e02382e85bb4d2d029973e36b96fb21569", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "4192046d9b914068c5692cda71f0194f7f3d499435552d0baefaf894355d07d1", "generation_seconds_batch": 25.114051705924794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3368031978607178, "target_z": -0.32421879708078466, "ce": 6.117925643920898, "genuineness": 4.268857479095459, "identity": 0.648505687713623, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/70be3f105134a2516d1db5251a03d3e02382e85bb4d2d029973e36b96fb21569.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "7127b8971be73c82246e8210d465e60a30e101c45d4b4f4e7484a972401751ac", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c510111e86825e67ba067dabd2272f8142e3c2b9bc5207add5957cd47a7df0f5", "generation_seconds_batch": 18.586013960884884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.374183177947998, "target_z": -0.2949218131380388, "ce": 6.046441555023193, "genuineness": 4.68307638168335, "identity": 0.7079446315765381, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/7127b8971be73c82246e8210d465e60a30e101c45d4b4f4e7484a972401751ac.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "7232fb8a276f8a697ea32f6beb87e52493d50f50b9474b8880f60b9031a5b01f", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "9f82229dce8d486b6fd1bffe6584e63896bd16b530b86b061be56e7feb8aa2c9", "generation_seconds_batch": 25.096467626048252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.524414539337158, "target_z": -0.9609376350379094, "ce": 6.268836498260498, "genuineness": 4.262581825256348, "identity": 0.2445349395275116, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/7232fb8a276f8a697ea32f6beb87e52493d50f50b9474b8880f60b9031a5b01f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "7234ff5dcd83c7afed066d8d3808b5af08e10846500caee2838a582644a2d5cb", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "9dd1b8fc841404ad18f11991776f53d8ea256f901cde7acff70563795a1685a1", "generation_seconds_batch": 22.547551158932038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.373470783233643, "target_z": 0.4882811970797382, "ce": 6.332377910614014, "genuineness": 4.645420074462891, "identity": 0.33207112550735474, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/7234ff5dcd83c7afed066d8d3808b5af08e10846500caee2838a582644a2d5cb.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "724768a60b0c64ed9d4d0b4fa52bb986c16ac2b99f232c365ed300795d649cfe", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "1173bd7350bd2cee0aab5ba7f8e9870dcfa8d94ef52a3da82f980b82580f7a1e", "generation_seconds_batch": 24.341178067028522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.723773956298828, "target_z": -0.8046874277364701, "ce": 6.006728172302246, "genuineness": 4.582659721374512, "identity": 0.7247973680496216, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/724768a60b0c64ed9d4d0b4fa52bb986c16ac2b99f232c365ed300795d649cfe.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "7248c8c3a02e2d41a6cfe4bfd16bb85714eebf08901b073cd0e5805d767290e2", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e75e01fc5d0a934398131e6df6d21606fb938fef130fc796ea13fefb2ba7dfec", "generation_seconds_batch": 19.89327829494141, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.129255771636963, "target_z": 0.2968749080282347, "ce": 6.022613525390625, "genuineness": 4.632867813110352, "identity": 0.7140288352966309, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/7248c8c3a02e2d41a6cfe4bfd16bb85714eebf08901b073cd0e5805d767290e2.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "727cc9c1cc955d0ad96cf357b3994122512496db8825b7392b15c687b862c45b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "a497bf9da7c3c6a245875c6cb48bdb44bb7c8b22ba3f84718beb769b8c264c65", "generation_seconds_batch": 19.78335841803346, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4551727771759033, "target_z": -0.23144529489032667, "ce": 6.006728172302246, "genuineness": 4.6705241203308105, "identity": 0.7604588270187378, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/727cc9c1cc955d0ad96cf357b3994122512496db8825b7392b15c687b862c45b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "72f11a1ce6447445e5b03e022a0f054dd96b0e68bd9e6d7027b429b32ba9048e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "be52715d19e45cca06b21537a343f4d02d81841225c586844858679344e135b6", "generation_seconds_batch": 21.121508384938352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.364215135574341, "target_z": -0.30273437956209154, "ce": 5.744620323181152, "genuineness": 4.049196243286133, "identity": 0.3874746561050415, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/72f11a1ce6447445e5b03e022a0f054dd96b0e68bd9e6d7027b429b32ba9048e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "7314ca093928292e208c71165021850cd250ce7e1ed3b98cef5051cfbc48148c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "b3fdf0f87e18d97512c1f6858927103acf799541216a5469ba442a7583ee2edf", "generation_seconds_batch": 22.328026499948464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.906756401062012, "target_z": 0.9062498569328095, "ce": 5.919358730316162, "genuineness": 4.406930446624756, "identity": 0.36334556341171265, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/7314ca093928292e208c71165021850cd250ce7e1ed3b98cef5051cfbc48148c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "73150f293408df9e3243b75718c1e741991ccbbb33d99f00047c49cbef5b8fd5", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "6bcc1c17ba1b17aa2a7940e60c02761383590fbb09775c307fa79948f9cba791", "generation_seconds_batch": 23.40896548796445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9929087162017822, "target_z": -0.5937500029197386, "ce": 6.006728172302246, "genuineness": 4.6956281661987305, "identity": 0.7780155539512634, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/73150f293408df9e3243b75718c1e741991ccbbb33d99f00047c49cbef5b8fd5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "749464ea42d0b7726e5c7c104406bba1c8c0f41f91b2b445a7cefef62b3c3d6d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "94f37430f6dd5f1fdc56bd62b99426e484b0b028a7323c6aaf4bbd4a781f1953", "generation_seconds_batch": 24.152745772036724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8563830852508545, "target_z": 0.08300784087620934, "ce": 6.022613525390625, "genuineness": 4.632867813110352, "identity": 0.7811559438705444, "hypothesis": "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_CASU_reg/749464ea42d0b7726e5c7c104406bba1c8c0f41f91b2b445a7cefef62b3c3d6d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "74ae0520796f590a68aa05134e88c51248363bcf344d5eaf6805b3c1794a6ba4", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "59c75380bd1bcc86415faaac8dd8f20e4dcaf656079bd8b90f47ab535c038f53", "generation_seconds_batch": 16.58148399204947, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.069447994232178, "target_z": 0.2499998832104567, "ce": 6.300607204437256, "genuineness": 4.469690799713135, "identity": 0.46547994017601013, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/74ae0520796f590a68aa05134e88c51248363bcf344d5eaf6805b3c1794a6ba4.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "751f17f9d878fdc67866dc05f6210e93728cf8f2c22357671cb5f5c80b053b32", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "33f9fbdb7e30a47fe9c67dc9de672b871bd4f7e3b8ac710a878a738cded0cccd", "generation_seconds_batch": 17.70816905994434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.933100938796997, "target_z": -0.6406250277375165, "ce": 5.879645347595215, "genuineness": 4.758388519287109, "identity": 0.7877030372619629, "hypothesis": "J'ai trouvé la 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_CASU_reg/751f17f9d878fdc67866dc05f6210e93728cf8f2c22357671cb5f5c80b053b32.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "7619456748dd940893cbc2f69bc51480a580ffee2ccbb6b78b5400371d63e48b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "bea5e3899ee70246e69cebc41e18dfecb5c561bb115c9b24ba5e12a95036921a", "generation_seconds_batch": 23.679947585915215, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.344991445541382, "target_z": -1.101562522627974, "ce": 5.839932441711426, "genuineness": 4.193545341491699, "identity": 0.4766691327095032, "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_CASU_reg/7619456748dd940893cbc2f69bc51480a580ffee2ccbb6b78b5400371d63e48b.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": "7635dbb795f83f8110d8c29fc8d5c034576472639cdf9f491a56abb393661c79", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1de1ed538b36106c079d9d1e6462d8686808d4bbf1536255d873a48e9de52236", "generation_seconds_batch": 23.274564691004343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5393667221069336, "target_z": -0.9492186919701957, "ce": 6.173524379730225, "genuineness": 4.3127899169921875, "identity": 0.6171306371688843, "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_CASU_reg/7635dbb795f83f8110d8c29fc8d5c034576472639cdf9f491a56abb393661c79.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "76a36b10198a4676fc657946ca5919da0925a695928b09440bd943cf66eac874", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "8d0b97e37b3642511939403ad06d309f3a62cf80d563da267ebf13ba310233d4", "generation_seconds_batch": 23.764447032939643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5536062717437744, "target_z": -0.1542969255479742, "ce": 6.09409761428833, "genuineness": 4.6203155517578125, "identity": 0.6038355827331543, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/76a36b10198a4676fc657946ca5919da0925a695928b09440bd943cf66eac874.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "772bfd5ad2408622586d9d3689a3b79b7de972df1f5c82170ff55f1f1a2c5c08", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3db43fc24568d4228176f61a9f1bc18afe1e7185c39a423dcfb002ce5ffc788d", "generation_seconds_batch": 22.79403879190795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.828437328338623, "target_z": -0.722656321168628, "ce": 6.141753673553467, "genuineness": 4.6203155517578125, "identity": 0.5912982225418091, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/772bfd5ad2408622586d9d3689a3b79b7de972df1f5c82170ff55f1f1a2c5c08.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "18f524694dd82814dedb56dcca1feaa51823fd572a4c259b553314eed89c3e69", "generation_seconds_batch": 16.51525559008587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3350234031677246, "target_z": -1.1093750890520266, "ce": 6.141753673553467, "genuineness": 4.494794845581055, "identity": 0.7624964714050293, "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_CASU_reg/77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "7841424839b9fabcdb4d4f3cbd36654bcebfe6ccfb315ced76d56489216aaada", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "cb1896b930584f4548049c296185361a19160225f0689f3660731d8a7060a955", "generation_seconds_batch": 22.54521551099606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.428294658660889, "target_z": 0.5312500321171244, "ce": 6.117925643920898, "genuineness": 4.557555198669434, "identity": 0.35503047704696655, "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_CASU_reg/7841424839b9fabcdb4d4f3cbd36654bcebfe6ccfb315ced76d56489216aaada.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "7894c6f3dd5a31488d0c1f9bf31e6d73efba8dfc19fff12680dcf6bd8d1ce365", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "d5188c242a5a4e092fee9fd6c4521d274ae13c4eb97a5e90deed9f2d00cf9f59", "generation_seconds_batch": 24.226769821951166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8035175800323486, "target_z": -0.7421874569338559, "ce": 6.109982967376709, "genuineness": 4.532451152801514, "identity": 0.6604255437850952, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/7894c6f3dd5a31488d0c1f9bf31e6d73efba8dfc19fff12680dcf6bd8d1ce365.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "78ce75ea8aec5d7c34bc50e2c8dfcf315e87b7a377aac342df01fcdb4daaad72", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ade09b4492090da0bdc1a10c4cc87b65e20088f85ee6b42b726e25571040c5d5", "generation_seconds_batch": 22.47733011993114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.498070240020752, "target_z": 0.5859374364956859, "ce": 5.645336627960205, "genuineness": 4.883909225463867, "identity": 0.7796180248260498, "hypothesis": "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_CASU_reg/78ce75ea8aec5d7c34bc50e2c8dfcf315e87b7a377aac342df01fcdb4daaad72.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "d032550367473497b546817cfba843ed35804cc8325c8f0aecd835a854e36416", "generation_seconds_batch": 16.936040572007187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4913065433502197, "target_z": -0.20312504525594802, "ce": 6.284721851348877, "genuineness": 4.632867813110352, "identity": 0.6515335440635681, "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_CASU_reg/794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "7a29f30240bb6e292b4bbbd6388473ec7e21bb1b07454f8bd18b5c1022e93f2a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "4a920126ecf9bab3df1d89834099255d6f98f5d35efffc12241aa00281aef537", "generation_seconds_batch": 22.41687322105281, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.914944887161255, "target_z": 0.12890631824888935, "ce": 5.609594821929932, "genuineness": 4.557555198669434, "identity": 0.7866337299346924, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Eh, dime cuándo estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/7a29f30240bb6e292b4bbbd6388473ec7e21bb1b07454f8bd18b5c1022e93f2a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "7a3f23c9b4d0dcbe29f774fa6129018e444153d41c8dbe273618c171ab1a701a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a4f8c0cb4bbd5f119c132f11d525bb6500191f753abdba48671d8c994939c15c", "generation_seconds_batch": 22.909954982926138, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1673479080200195, "target_z": -0.45703130511006573, "ce": 6.125868320465088, "genuineness": 4.093128204345703, "identity": 0.6288948059082031, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/7a3f23c9b4d0dcbe29f774fa6129018e444153d41c8dbe273618c171ab1a701a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "7b2438b86a6f230df9ed7b9895d2a870328b852b9a75e88e0fa28854c08856a7", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "2b65205ab5e85a27c6d0a1deb2f4c442e890627ac6563852e5d71e5d0a581fdd", "generation_seconds_batch": 18.586013960884884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3592312335968018, "target_z": -0.3066405693424833, "ce": 5.967014789581299, "genuineness": 4.6956281661987305, "identity": 0.792130708694458, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/7b2438b86a6f230df9ed7b9895d2a870328b852b9a75e88e0fa28854c08856a7.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "7b42f881239b4e2ed4a30e29bfac2c05f73b2c282bffb54bb30be6999a8b91e9", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "feb731a9985b2aec452eeb85dd06ae348cfa2fdd818ac8e51282e68acd225861", "generation_seconds_batch": 22.66537852305919, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.745177984237671, "target_z": -0.004150420073300854, "ce": 6.078212261199951, "genuineness": 4.381826400756836, "identity": 0.7308269739151001, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/7b42f881239b4e2ed4a30e29bfac2c05f73b2c282bffb54bb30be6999a8b91e9.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "7b4b1028dd22a84395f331c2109074ed426719600ecef8dfb7c21a2315be1b21", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "83044cdbe2e5f8ee5fc3cde1afd4b062f8d0844c1b2e4ed5581d38367daa87c8", "generation_seconds_batch": 16.734484044951387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.58778190612793, "target_z": 0.6562499737223527, "ce": 6.054384231567383, "genuineness": 4.494794845581055, "identity": 0.5940488576889038, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/7b4b1028dd22a84395f331c2109074ed426719600ecef8dfb7c21a2315be1b21.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "7b966ab5149cca4d95113ba7d86982b5210455997dbf28d0b67d02a1817815d7", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4a9a89627f743f2169cc23a4c18b5765f8d1cbcd42c7fd3a02e10655a9d2f19c", "generation_seconds_batch": 24.038956668111496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.08262038230896, "target_z": -0.5234374656930717, "ce": 6.221180438995361, "genuineness": 4.632867813110352, "identity": 0.7467775344848633, "hypothesis": "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_CASU_reg/7b966ab5149cca4d95113ba7d86982b5210455997dbf28d0b67d02a1817815d7.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "7ccfd25f2564f6ab4470a50ef2aecdb7fa707081ccfc5dab1fad48396e1396fc", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "4c5970d955fcd448b996ef55ef4fbfb6511597a571e8b84b11bc99b831b8c37f", "generation_seconds_batch": 16.70463887299411, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.635219097137451, "target_z": -0.09033194671444386, "ce": 5.990842819213867, "genuineness": 4.65797233581543, "identity": 0.8483743071556091, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/7ccfd25f2564f6ab4470a50ef2aecdb7fa707081ccfc5dab1fad48396e1396fc.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "7db92134bdada12e2dc6eac897b8934ea8de777edb9637a99dd5f544abfdd276", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "431de61f5ed776bc90e65667fbe87342fc5c2c142524ec629117375026cf70ad", "generation_seconds_batch": 22.208209252916276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.356739044189453, "target_z": -0.3085938510959484, "ce": 6.046441555023193, "genuineness": 4.469690799713135, "identity": 0.618903636932373, "hypothesis": "La puerta sigue abierta y he dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/7db92134bdada12e2dc6eac897b8934ea8de777edb9637a99dd5f544abfdd276.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "7e5191d667d48abf7cc149f099e7b1e8068881d39c25b0f2640997988e406c95", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "79cc40840618de454e3b7eb68d0198790b3f8fa8b64fcf46f3e51f7b748b9542", "generation_seconds_batch": 22.108859014930204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.926692485809326, "target_z": 0.921874989780915, "ce": 6.332377910614014, "genuineness": 4.570107460021973, "identity": 0.304984986782074, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/7e5191d667d48abf7cc149f099e7b1e8068881d39c25b0f2640997988e406c95.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "7e7245d559c3a33582a2c472848594ce5e48f3dd4b9872a73143fe52b422c888", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5bf5c8c38364c7053fa0e2feacd6be49cf048bb5f95a842707dabb748e2eb29d", "generation_seconds_batch": 23.73444573301822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.615906000137329, "target_z": -0.10546880584000039, "ce": 5.764476776123047, "genuineness": 4.65797233581543, "identity": 0.653692364692688, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/7e7245d559c3a33582a2c472848594ce5e48f3dd4b9872a73143fe52b422c888.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "7ec4a83431ca92d67b27167695ad6b746db033ad68bd0a51248751f106cabbf6", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "7957ee88048ac94d901b680bcdb118f31dcbdb5dded500c215038281b4cb3525", "generation_seconds_batch": 21.43731519393623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7040600776672363, "target_z": -0.03637704635134051, "ce": 6.133810997009277, "genuineness": 4.532451152801514, "identity": 0.6402992606163025, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/7ec4a83431ca92d67b27167695ad6b746db033ad68bd0a51248751f106cabbf6.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "7ec7c5eac6a4d4a7788755cc2dd46b75f87e002d771f6291dd6e1d8c2d024301", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "240f36af32fc400424be4609645b4f38566348383ea9408f8018d200089a97f8", "generation_seconds_batch": 16.7550384439528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3368031978607178, "target_z": -0.32421879708078466, "ce": 5.744620323181152, "genuineness": 4.7332844734191895, "identity": 0.6539965867996216, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/7ec7c5eac6a4d4a7788755cc2dd46b75f87e002d771f6291dd6e1d8c2d024301.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "7f013c41996b2398b3a992be42054a77270dd5e97bf8dc0d273b1feb07e3b348", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "55f69ec351f53b39a4edf5d011d3ea3bd6d85a6f04565a4b37a60f4713a9ac1c", "generation_seconds_batch": 22.151396099012345, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.283759117126465, "target_z": 0.4179686598530713, "ce": 5.847874641418457, "genuineness": 4.7081804275512695, "identity": 0.7081711292266846, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/7f013c41996b2398b3a992be42054a77270dd5e97bf8dc0d273b1feb07e3b348.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "ad09bbf9ab10fa699472017dbed845ec5fe326d3f6f0acc6331574c7973c03d3", "generation_seconds_batch": 17.3396077939542, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4987826347351074, "target_z": -0.19726557372209116, "ce": 6.507116317749023, "genuineness": 4.507347106933594, "identity": 0.4615280330181122, "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_CASU_reg/7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "7fa6510c66e5fcbfaf84a3b273a652a07209ae03010e738658421d1c18caa421", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "60fa0485538a8f9dc2b1a49ad0aa75fbe671afbde8df85a6649a184c4a410dbd", "generation_seconds_batch": 22.333664865931496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.218967437744141, "target_z": 0.3671874452549016, "ce": 5.669164657592773, "genuineness": 4.482243061065674, "identity": 0.7785370945930481, "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_CASU_reg/7fa6510c66e5fcbfaf84a3b273a652a07209ae03010e738658421d1c18caa421.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "803ee25130025184b1d2fd6382b4ecfaffc9b5aa2f620bee402b6461d68cfee5", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "ebb681d563b53d8ad52f941456dfbac486d9116d7a7984e95bbcb3ebb699b016", "generation_seconds_batch": 17.446140857995488, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.154175758361816, "target_z": 0.3164062306567319, "ce": 6.0146708488464355, "genuineness": 4.256305694580078, "identity": 0.6584525108337402, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen samen kijken.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/803ee25130025184b1d2fd6382b4ecfaffc9b5aa2f620bee402b6461d68cfee5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "8055dee56b2955087a26fc18329522d15c91e9243d2f5762763a7c560301bf0e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e15b3c38184488b4720775b5d3d4e8a319c3310a1d9349c33ff3aa1388735789", "generation_seconds_batch": 22.906182117993012, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.551114320755005, "target_z": -0.15625002043817007, "ce": 6.284721851348877, "genuineness": 4.356721878051758, "identity": 0.27789944410324097, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8055dee56b2955087a26fc18329522d15c91e9243d2f5762763a7c560301bf0e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "80ad9a6bd022419d955ac5326a87cf199eb1ad298e95fcd1bad1fabdd97ea043", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b0a2aa9bd8e0533429e457c33b6319e68c61ca17fc0c57fe24ea9dc08fd9e11", "generation_seconds_batch": 21.743965385016054, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.383439064025879, "target_z": 0.49609395036706017, "ce": 5.895530700683594, "genuineness": 4.570107460021973, "identity": 0.2641165852546692, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/80ad9a6bd022419d955ac5326a87cf199eb1ad298e95fcd1bad1fabdd97ea043.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "80bd36406a2bfaaea25252f29d975da42dd43f045482c35765283397879a0234", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "36de1da603a17ef7760da3050f67e1a44c8c04e50214b701856ddf152116c165", "generation_seconds_batch": 22.826197736081667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.729914665222168, "target_z": -0.0161132197073852, "ce": 6.070269584655762, "genuineness": 4.65797233581543, "identity": 0.7111532092094421, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/80bd36406a2bfaaea25252f29d975da42dd43f045482c35765283397879a0234.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "80d90b2040bb45d590ca92b7cc22b5263d0b22c618ed1e695a50d3693464b164", "generation_seconds_batch": 17.087034616968594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8533573150634766, "target_z": -0.7031249985401307, "ce": 6.149696350097656, "genuineness": 4.20609712600708, "identity": 0.6480227112770081, "hypothesis": "Las llaves están en la mesa de la cocina. Miren el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.08695652173913043, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "80ce7a1c670eacee014c64d91ef47739c3c2a6f7a69fa06df726fd6cfd14cee0", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "b02f9e67540475211ffec1dc7f9e60a62653561ab1ba1be45104de7021133fca", "generation_seconds_batch": 23.588325946009718, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.3635029792785645, "target_z": 0.4804688175189547, "ce": 5.704906940460205, "genuineness": 4.545003414154053, "identity": 0.5348697900772095, "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_CASU_reg/80ce7a1c670eacee014c64d91ef47739c3c2a6f7a69fa06df726fd6cfd14cee0.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "81064a53ebbbc8006874cd06c6932c14f33d8473e6020fd1a27d1aff37f78760", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f0e666199f170d6438c82c509da7fe3450e9b272dc7f28019bc5f51bfb7259e1", "generation_seconds_batch": 22.288060329970904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8688430786132812, "target_z": 0.09277350219045796, "ce": 5.77639102935791, "genuineness": 4.375550270080566, "identity": 0.13051125407218933, "hypothesis": "Ik heb het een kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/81064a53ebbbc8006874cd06c6932c14f33d8473e6020fd1a27d1aff37f78760.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "8168b23617ef01f2102e2e534e3e43f2f8c76b08799df7cba1001cfb441aeca8", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "912a098b9885cb9958f55ee265edaf8cdf833641ed026f4bf6b3973e65ad1809", "generation_seconds_batch": 23.679947585915215, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1349520683288574, "target_z": -0.48242191240915056, "ce": 5.919358730316162, "genuineness": 4.300237655639648, "identity": 0.6311289072036743, "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_CASU_reg/8168b23617ef01f2102e2e534e3e43f2f8c76b08799df7cba1001cfb441aeca8.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": "81835bf47becd83193942ffe7274f17c20cccc0537e92fa2c10c0d30b0519df6", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3489634e09807dd142eebcf796c0cf89d67541a3b0bf9941026fa657534ebaa5", "generation_seconds_batch": 24.24899140105117, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.303695201873779, "target_z": 0.43359379270117676, "ce": 5.847874641418457, "genuineness": 4.400654315948486, "identity": 0.26356932520866394, "hypothesis": "Sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.045454545454545456, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/81835bf47becd83193942ffe7274f17c20cccc0537e92fa2c10c0d30b0519df6.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "818c91814427dda7ee26a13f041498dcd6ecc89f32f4e593d1bbddd535cd0048", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c4e1a8b15662461f0c2196fbee75dc0e8ee3c84a99a5415df6ff5e143d48894d", "generation_seconds_batch": 21.547154724015854, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3617231845855713, "target_z": -0.3046874744522874, "ce": 6.038498878479004, "genuineness": 4.7081804275512695, "identity": 0.7785701751708984, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/818c91814427dda7ee26a13f041498dcd6ecc89f32f4e593d1bbddd535cd0048.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "81a00430549cfb2a02f81b34ab4d1b0c7eb2f9cec7dba3539c054a7ab9af9596", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "d0bae2247a266bdc1e0ff8058b88fc4a80e19535277974ecc2be75d33a9cb8d9", "generation_seconds_batch": 25.43309945904184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4745750427246094, "target_z": -0.9999999065683653, "ce": 6.356205940246582, "genuineness": 4.507347106933594, "identity": 0.595321774482727, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/81a00430549cfb2a02f81b34ab4d1b0c7eb2f9cec7dba3539c054a7ab9af9596.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "81b2fa08863a62b92073ed4c23b150ba9ef4f3bff2ac6c590def86a41e205910", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "d67c6f58acca35786d510e0652561fe75328b2be6f02b04f35c26935f455dab5", "generation_seconds_batch": 16.27507096901536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9562411308288574, "target_z": -1.4062499970802613, "ce": 6.419747352600098, "genuineness": 4.13078498840332, "identity": 0.563176155090332, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/81b2fa08863a62b92073ed4c23b150ba9ef4f3bff2ac6c590def86a41e205910.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": "8339876f37302cc7534bf2f64442727e90e864298db28e0c38147fe5dd1dd1f1", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b64915dd7c61f94cbba8bc8eccb3387bba008245d34e9e0c0b853c887ae3067", "generation_seconds_batch": 18.27532177802641, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.728668689727783, "target_z": -0.017089767152483136, "ce": 6.0305562019348145, "genuineness": 4.65797233581543, "identity": 0.6448296308517456, "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_CASU_reg/8339876f37302cc7534bf2f64442727e90e864298db28e0c38147fe5dd1dd1f1.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "836a30a1432ad994a0976a7868287b9e1db0d9a5d8b8ab7e99923d55cf5c70fd", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "cb3e591c98211b73f94a5506e637d8e8e2f04403169d8c13dc63fbb271b64bd1", "generation_seconds_batch": 22.848537919926457, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.248871326446533, "target_z": 0.3906249576637906, "ce": 5.831989288330078, "genuineness": 4.720732688903809, "identity": 0.6298871040344238, "hypothesis": "Me he encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/836a30a1432ad994a0976a7868287b9e1db0d9a5d8b8ab7e99923d55cf5c70fd.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "837cb42aaacbb8c4f6af590aa53d61bb4a6a5a5290bf8a42516bd0ab8955ef9b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "1257d8b695f6d089b5d321f5deb1345cb85b65a921110bf9a4db08f257aebf45", "generation_seconds_batch": 23.93859411496669, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.428294658660889, "target_z": 0.5312500321171244, "ce": 5.974957466125488, "genuineness": 4.6203155517578125, "identity": 0.5307859182357788, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt ?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/837cb42aaacbb8c4f6af590aa53d61bb4a6a5a5290bf8a42516bd0ab8955ef9b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "843d6d9a1c4a9f03d988e18b07caaf5727e81e0d05c3b5e2a73d91a9c7734936", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "834f5f591c3758fcb7082e6b4cccb6c928838c881d52f019a7c3ca68d8634123", "generation_seconds_batch": 22.416251708986238, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.647589683532715, "target_z": 0.7031249985401307, "ce": 6.09409761428833, "genuineness": 4.570107460021973, "identity": 0.4854891896247864, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/843d6d9a1c4a9f03d988e18b07caaf5727e81e0d05c3b5e2a73d91a9c7734936.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "85006674aedef1da906e3046c2e35d380927e21c31b2037d68014af21d7392a2", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "9093510f1a57c09fe21448ea277f15d624a5c46c2f255b0d711962804cb27dc7", "generation_seconds_batch": 22.70243758999277, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9961129426956177, "target_z": -1.3750000116789542, "ce": 6.316492557525635, "genuineness": 4.344170093536377, "identity": 0.40386706590652466, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/85006674aedef1da906e3046c2e35d380927e21c31b2037d68014af21d7392a2.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "8588b4c5fbee69e175b48c35f1c10299bcd2c8ff5a0e4287b12c48b31dbf30da", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e507c45e60b84a97a4903824a4cd386e4c8fe8aca3f6b0ce26445e508188936a", "generation_seconds_batch": 26.236191531992517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.734898567199707, "target_z": -0.012207029926993455, "ce": 6.205295085906982, "genuineness": 4.300237655639648, "identity": 0.4252915382385254, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8588b4c5fbee69e175b48c35f1c10299bcd2c8ff5a0e4287b12c48b31dbf30da.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "85e0c2820a4409a4d9bff768ff4c6407f110b3f02808dd4ac51bbb1468dad9fa", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "6b975dfab8e466caf42be256e2c102231a31797db330afc95fdc92755d1d0360", "generation_seconds_batch": 16.04218107496854, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.3684868812561035, "target_z": 0.48437500729934646, "ce": 6.062326908111572, "genuineness": 4.595211505889893, "identity": 0.6511842608451843, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/85e0c2820a4409a4d9bff768ff4c6407f110b3f02808dd4ac51bbb1468dad9fa.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "864f2bd267c2d53e0ad8d205ec05fb2c26412bc10ef21d06fc56d9b89d02da13", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1508e0d6dd95327640a48a65536960152edbd5841dcb568cdc0b0e11eb2b23ab", "generation_seconds_batch": 23.29694223497063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3118834495544434, "target_z": -0.34374993284601263, "ce": 5.780362129211426, "genuineness": 4.557555198669434, "identity": 0.6500991582870483, "hypothesis": "Resie Serra Mamba de Moy shouldn't be. She troubled the petite boat, I tell the fenêtre. We got ensemble what container.", "wer": 1.105263157894737, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/864f2bd267c2d53e0ad8d205ec05fb2c26412bc10ef21d06fc56d9b89d02da13.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "86c63f4da23fec8d6f2caf206104c2174fe6a57ff52d15c1dfac1ec87b2b6bd7", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "351c52ed2ab24d33973ab91141bc7b2974c69c61301be9d394893f8d0e1d54a3", "generation_seconds_batch": 21.483801538008265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2353439331054688, "target_z": -1.1875000058394771, "ce": 6.419747352600098, "genuineness": 4.457139015197754, "identity": 0.47128045558929443, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/86c63f4da23fec8d6f2caf206104c2174fe6a57ff52d15c1dfac1ec87b2b6bd7.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "86f2d513f3fb2a128fe8d5b74683731e8e28206b3117ada3c7f4ecc1ecfd9f96", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d889625665d07cf700578ee0bb564b1dcae2a8e71e8d8b082d9be0dfc34df16f", "generation_seconds_batch": 23.130981167894788, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.404799222946167, "target_z": -1.054687497810196, "ce": 5.990842819213867, "genuineness": 3.8295347690582275, "identity": 0.3874053359031677, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/86f2d513f3fb2a128fe8d5b74683731e8e28206b3117ada3c7f4ecc1ecfd9f96.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "8730b36969b7cb85080953a7b31763424514775c5b0086addaedd1d1ab548505", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "b19bf11c81f05c89c18a9636125d5ade83f8eb66613e106a3729ea6ac31be704", "generation_seconds_batch": 24.226769821951166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.977956771850586, "target_z": -0.6054687591241831, "ce": 6.141753673553467, "genuineness": 4.632867813110352, "identity": 0.5161870121955872, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8730b36969b7cb85080953a7b31763424514775c5b0086addaedd1d1ab548505.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "887852e60dce60d23d8098d9d6fa1ae0aec9b004639756e2f89f51ffa5a7a561", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "e868c40892e953d574230686bc3c7d0f6612d39080ad325006282fd324b99187", "generation_seconds_batch": 24.038956668111496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5728302001953125, "target_z": 0.6445314043811775, "ce": 6.09409761428833, "genuineness": 4.6956281661987305, "identity": 0.7628167867660522, "hypothesis": "Te he encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/887852e60dce60d23d8098d9d6fa1ae0aec9b004639756e2f89f51ffa5a7a561.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": "88e010b9d9234aacc14e524c9538bbd2f83f6f6b56fa497933bc157d942bfec1", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Your message arrived after I had left.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f53200333f198a06edcd4efe338e058e4e35ce11dfe8ff95164dca7617f13f8c", "generation_seconds_batch": 18.10422323900275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4701247215270996, "target_z": -0.21972653868588218, "ce": 6.205295085906982, "genuineness": 4.18099308013916, "identity": 0.4345153570175171, "hypothesis": "Your m message arrived after I had left. I r read it twice before putting the phone away. Let me know when we can talk.", "wer": 0.08695652173913043, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/88e010b9d9234aacc14e524c9538bbd2f83f6f6b56fa497933bc157d942bfec1.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "88f0e21e3a903f2ea0f070e8a0ee932483f96a20f8ab3f881746e73b0e28cac3", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "cf1c5533b0d68d0229fb28950280cb7fc67efce1d693d747f055ca080b4c9db9", "generation_seconds_batch": 15.876746883965097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.237123727798462, "target_z": -0.40234371386823503, "ce": 6.133810997009277, "genuineness": 4.557555198669434, "identity": 0.6609866619110107, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/88f0e21e3a903f2ea0f070e8a0ee932483f96a20f8ab3f881746e73b0e28cac3.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "88fca4a64d069383401bfd6471b16510ccec5e402c319fe154a5d7b5f81a18c6", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "46af4fb3517ccc81c79152710b51aa42e889ad30738bb0b459a4c883bd4c3d9e", "generation_seconds_batch": 23.575429996009916, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0726523399353027, "target_z": -0.5312500321171244, "ce": 6.260893821716309, "genuineness": 4.3881025314331055, "identity": 0.6358739137649536, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/88fca4a64d069383401bfd6471b16510ccec5e402c319fe154a5d7b5f81a18c6.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "895d6abdcffc251792febdfbe064942dbe857896938b5692de648a176f66b675", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "1505709d0a9a3b2533d5615fa6ae9565305b08c975fe06ef486ab96ed4b65ff3", "generation_seconds_batch": 16.840930305072106, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4745750427246094, "target_z": -0.9999999065683653, "ce": 6.1894097328186035, "genuineness": 4.18726921081543, "identity": 0.29862135648727417, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/895d6abdcffc251792febdfbe064942dbe857896938b5692de648a176f66b675.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "89a49267388c18b4cc823254af170b79141bf0a875ac4d4cdd0c5fe76d23e83f", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "33bc75ac719060b6f5d388cb412adf5ad2e10c3546dae97a83728c1c8d9d6f8f", "generation_seconds_batch": 17.064507020055316, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.354959487915039, "target_z": -1.0937499562039212, "ce": 6.300607204437256, "genuineness": 4.21237325668335, "identity": 0.3460071384906769, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/89a49267388c18b4cc823254af170b79141bf0a875ac4d4cdd0c5fe76d23e83f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "8a243b13d5ab891270a24935832c2a94d80e40a5227dd0aa7d3eda3d1c3efa08", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "d140750b7e3baac627b51e7993a91d797099bf8c8570ecad94752994ce85133c", "generation_seconds_batch": 22.28225635702256, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7230615615844727, "target_z": -0.0214844175186931, "ce": 6.038498878479004, "genuineness": 4.375550270080566, "identity": 0.24851906299591064, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8a243b13d5ab891270a24935832c2a94d80e40a5227dd0aa7d3eda3d1c3efa08.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "8a43f920c358b7dc17e3636ca95e746c2d71e51a8499406ecffe6f85acb9650a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "9983a346253f9981297e972f8591b0ddf2373fd979932209d0686172698969b9", "generation_seconds_batch": 17.00756287202239, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.856916904449463, "target_z": 0.8671875854023535, "ce": 5.935244083404541, "genuineness": 4.582659721374512, "identity": 0.5445376634597778, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8a43f920c358b7dc17e3636ca95e746c2d71e51a8499406ecffe6f85acb9650a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "8a5f6d68c86b42725ac21bd429bb7a466ca87615c79aaf7b2bc5d3534f8a11ce", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "54333be1df84f923e62237bbcbf9b3d517ced875baef92655d0799789a7a0d8d", "generation_seconds_batch": 22.54521551099606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.226443290710449, "target_z": 0.3730467299254892, "ce": 5.911416053771973, "genuineness": 4.557555198669434, "identity": 0.4108654260635376, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8a5f6d68c86b42725ac21bd429bb7a466ca87615c79aaf7b2bc5d3534f8a11ce.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": "8a73902e01c77542345d3f673f372edc2dddfff0402f02e7a3b0fbb820251ee3", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "11f9b85a5d19333a2e16991107382d1acbfa528af5617846a453dc65e6b69115", "generation_seconds_batch": 17.345418073935434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.794706344604492, "target_z": 0.034667994890784495, "ce": 5.7247633934021, "genuineness": 4.6203155517578125, "identity": 0.8020790219306946, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, le bien, et vérifié les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.09090909090909091, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8a73902e01c77542345d3f673f372edc2dddfff0402f02e7a3b0fbb820251ee3.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": "8a797d37fd1a33b6ea299543c7c7e4b6052f381b133f411d178a2df8099f2cc5", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3eb7734b4eca99f5004b0e9791297ccddf4c3a494f717c606da3f909b1519ae", "generation_seconds_batch": 19.861331301974133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.91369891166687, "target_z": 0.12792977080379142, "ce": 5.831989288330078, "genuineness": 4.595211505889893, "identity": 0.5643475651741028, "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_CASU_reg/8a797d37fd1a33b6ea299543c7c7e4b6052f381b133f411d178a2df8099f2cc5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "8ac21028afaf3734bbc1f4276ca08b82e9115eff882233a6018e94aab18a01d0", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "cdb1d3a58c37aacdab4cd8a37839c7b84ac18b1572ec7569ad4c678ed84ab593", "generation_seconds_batch": 23.679833487025462, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.388422966003418, "target_z": 0.500000140147452, "ce": 5.95112943649292, "genuineness": 4.8588056564331055, "identity": 0.639647901058197, "hypothesis": "He encontrado la cajita, estaba junto a la a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.11764705882352941, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8ac21028afaf3734bbc1f4276ca08b82e9115eff882233a6018e94aab18a01d0.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "8b2534616cddce5ae305a8754c516df676f0781531751d6265d7d305bff9a1e7", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "308c8138099f68ad748db5029f4092ab868cc8a8287a028ed1100d91287e3919", "generation_seconds_batch": 16.734484044951387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3848633766174316, "target_z": -1.0703124437950322, "ce": 6.117925643920898, "genuineness": 4.256305694580078, "identity": 0.6351445913314819, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8b2534616cddce5ae305a8754c516df676f0781531751d6265d7d305bff9a1e7.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "8b2a0163a68a1c73f65b236bcf1a9a2f66679f720acb47fdfa062fc4c6a41eb0", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "394d0efc988c6b5cec9ab38e0fcc818a531d74a25e72085c9b2b4ae4d1fc8e28", "generation_seconds_batch": 22.593999217962846, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.268807411193848, "target_z": 0.40625009051189603, "ce": 5.784333229064941, "genuineness": 4.582659721374512, "identity": 0.8224846124649048, "hypothesis": "Le la porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8b2a0163a68a1c73f65b236bcf1a9a2f66679f720acb47fdfa062fc4c6a41eb0.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": "8b31baaa395b825b71a8b4cbabe866874655bc7b287f9fc66089f59204ab020c", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I wrote the number on the envelope.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6e8f58c1b1174d8e9234f93c213629dadecdbbb2d5a54745094a957849c4bcce", "generation_seconds_batch": 17.15580042800866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.291234970092773, "target_z": 0.4238279445236589, "ce": 6.221180438995361, "genuineness": 4.6203155517578125, "identity": 0.5819488763809204, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8b31baaa395b825b71a8b4cbabe866874655bc7b287f9fc66089f59204ab020c.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": "8b4843a37f998847ec18d261ac63c3365ced4782404ed0f49c6802ab00f52add", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e14566fa2b087c2f118a1f4b6e03c12434dc59af0b6610ca337e9915e3f2a6b", "generation_seconds_batch": 17.60428318497725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.772189140319824, "target_z": 0.8007812379560784, "ce": 5.919358730316162, "genuineness": 4.432034492492676, "identity": 0.4361787736415863, "hypothesis": "De weg gaat linksaf naar de brug. Loopt verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8b4843a37f998847ec18d261ac63c3365ced4782404ed0f49c6802ab00f52add.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "8b72061ba2faa795db118d8cd2036c05bdfebe4ea1bae916274f4edd506153b5", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "f9bd5e446ab3dc761d6494373c6484cf7c5974dcd0712f3085fda1c9cbef0a94", "generation_seconds_batch": 22.73331608402077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.04274845123291, "target_z": -0.5546875445260133, "ce": 6.09409761428833, "genuineness": 4.6203155517578125, "identity": 0.7056885957717896, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi dis-moi quand tu es prêt.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8b72061ba2faa795db118d8cd2036c05bdfebe4ea1bae916274f4edd506153b5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "8bc4e551959c731d75725aed1c1b2e80179affc563bf3f664246a0dd2cd3ef22", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e629f08240ab9337af6ea08d556274be3c9ddfa1f7799317d22ea46d8db3fbac", "generation_seconds_batch": 16.921104872017168, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.252075672149658, "target_z": -0.3906249576637906, "ce": 5.780362129211426, "genuineness": 4.720732688903809, "identity": 0.75678551197052, "hypothesis": "J'ai trouvé la petite boîte , elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8bc4e551959c731d75725aed1c1b2e80179affc563bf3f664246a0dd2cd3ef22.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": "8bd25cb540b722712bdd90f499fa55b3b1aa724ad76db1ce190ab262a8297349", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "43da8565d753025957ee06f742bdc4bc060e8de67c793f2404f69e05bfe01b95", "generation_seconds_batch": 16.14351812598761, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6757137775421143, "target_z": -0.05859378102222244, "ce": 5.903473377227783, "genuineness": 4.632867813110352, "identity": 0.6422213912010193, "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_CASU_reg/8bd25cb540b722712bdd90f499fa55b3b1aa724ad76db1ce190ab262a8297349.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "8bff6cc54c1e1060ffd7aba6a9b73f85850e079ea5bd036b65105ec4da919974", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9e04ef4de33686e4ef7bd7eeab928a85324f6d36ee6a1ec81f5deca2ac1ec094", "generation_seconds_batch": 23.137088172021322, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.08262038230896, "target_z": -0.5234374656930717, "ce": 6.086154937744141, "genuineness": 4.7081804275512695, "identity": 0.5246957540512085, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8bff6cc54c1e1060ffd7aba6a9b73f85850e079ea5bd036b65105ec4da919974.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "8c44113b9cb73052b4895b7b557098d8e092cd241d496a90a30c4289f847d772", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "0484b8ecf76cba101526bd38bfd86133bced7c31b31d7868fc468aff3ce3695f", "generation_seconds_batch": 24.152745772036724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5536062717437744, "target_z": -0.1542969255479742, "ce": 5.990842819213867, "genuineness": 4.494794845581055, "identity": 0.649568498134613, "hypothesis": "Me he encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8c44113b9cb73052b4895b7b557098d8e092cd241d496a90a30c4289f847d772.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "8c5e8c797e906b72a0ef98ddd2deaa5a06767c7165290e0f5d2ca31ff4b4b88b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "0bd866a55a3bc88cae74c96cf8080aa933aad71bec5f688fdfe14a9086e4aeee", "generation_seconds_batch": 21.95207603799645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1865715980529785, "target_z": 0.3417968379558167, "ce": 5.935244083404541, "genuineness": 4.758388519287109, "identity": 0.7422041893005371, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8c5e8c797e906b72a0ef98ddd2deaa5a06767c7165290e0f5d2ca31ff4b4b88b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "8cb354c448f024018185dd2cf742c92a7645c0963928106d717dcc8f26ba2f82", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "577fb36018e9bac7ad2bac0068336683b4bc0d2fee0b20afb2c38cb6353916cb", "generation_seconds_batch": 25.03083440696355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.462648630142212, "target_z": -0.22558601021973906, "ce": 6.0305562019348145, "genuineness": 4.256305694580078, "identity": 0.630171537399292, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8cb354c448f024018185dd2cf742c92a7645c0963928106d717dcc8f26ba2f82.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "8d8b8ffd0606cfb0937855ce5c5a75c17b4746909c118af2ffa747888e839625", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6682656c78b5e73fa011c690084842fc0fbfeedc7edd9cb8f04c7dac052c2da0", "generation_seconds_batch": 21.827974291983992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9722604751586914, "target_z": 0.1738280613132022, "ce": 6.086154937744141, "genuineness": 4.482243061065674, "identity": 0.4717257022857666, "hypothesis": "The door is s still open. I put your coat on the chair. Tell me when you are ready.", "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_CASU_reg/8d8b8ffd0606cfb0937855ce5c5a75c17b4746909c118af2ffa747888e839625.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "8dd6f7e4880677ebadcd4934f73d30e340f1c49b865542c18bb461675d12792e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "66790498f41df7921ebdede359597df39e56dfea3ac4179f2627acfc701ec8df", "generation_seconds_batch": 22.556334809982218, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7835817337036133, "target_z": -0.7578124029186921, "ce": 6.284721851348877, "genuineness": 4.469690799713135, "identity": 0.6418741941452026, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8dd6f7e4880677ebadcd4934f73d30e340f1c49b865542c18bb461675d12792e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "8e4bebad59d5dbde0c964eb081ca13b70d1e1c9c6a4048dea914868409b46262", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e1fed99e6c35bebf8313b34fbe176c2bea314abcab664bb47e7522527e8b5e79", "generation_seconds_batch": 16.39809112693183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7741475105285645, "target_z": 0.01855477518339929, "ce": 6.173524379730225, "genuineness": 4.7458367347717285, "identity": 0.7062839865684509, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es quand tu es prêt.", "wer": 0.16666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8e4bebad59d5dbde0c964eb081ca13b70d1e1c9c6a4048dea914868409b46262.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "8ea439bcbe381f1e88dc1ede2a5490f7022d7f6e4c28c70df22f2ef76ac22eae", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c51eed0f62d4a327e8a357fdf033baa1fc4ca0e4452396ae73b3d085818c5f73", "generation_seconds_batch": 24.960537275997922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9380850791931152, "target_z": -0.6367186510938555, "ce": 5.998785495758057, "genuineness": 4.595211505889893, "identity": 0.6776835918426514, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8ea439bcbe381f1e88dc1ede2a5490f7022d7f6e4c28c70df22f2ef76ac22eae.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "8f45472b6d6b7c6f469aac5a6397ecc323d65226ee2faa1d8baffc16bc2b11d2", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "cef83ce7cd66919bb8ad941c154bf307f8c61353371d6231078e312962196619", "generation_seconds_batch": 22.681499610072933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.668950080871582, "target_z": -0.8476562627738563, "ce": 6.006728172302246, "genuineness": 4.18099308013916, "identity": 0.7758708000183105, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8f45472b6d6b7c6f469aac5a6397ecc323d65226ee2faa1d8baffc16bc2b11d2.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "8f782849872e9c03442cea44e8835cd6abfeb8190ad645c666ac65deb26dd2ae", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "804a0c42ce5233237b578c7ef704531657342014871d213527ff7535f94e5a3b", "generation_seconds_batch": 16.846680309972726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7967537641525269, "target_z": -1.5312500321171243, "ce": 6.2767791748046875, "genuineness": 4.300237655639648, "identity": 0.6700515747070312, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/8f782849872e9c03442cea44e8835cd6abfeb8190ad645c666ac65deb26dd2ae.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "908d7fd68ab71df899fbbb69869f22fa3acdb468220e1dd9466dcbf07aaebb4d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2a81b9832ecadd9005bcd421de22571b988c9ba79f73305a9810aa764e3c52f5", "generation_seconds_batch": 23.536171647021547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.071939945220947, "target_z": 0.2519529781006526, "ce": 6.062326908111572, "genuineness": 4.325342178344727, "identity": 0.1667540818452835, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/908d7fd68ab71df899fbbb69869f22fa3acdb468220e1dd9466dcbf07aaebb4d.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": "91033af3f492334efe5bda6dcb8338ba25a04bdb2de3adcf8d45c7b8eaab0651", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c68779080b3daf1a5bf560c082e60e333f45ad1869a68e996b142e5d8a131e43", "generation_seconds_batch": 22.728559091105126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.607807159423828, "target_z": -0.11181636423313697, "ce": 5.824047088623047, "genuineness": 4.796044826507568, "identity": 0.7488603591918945, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue, elle était ici près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/91033af3f492334efe5bda6dcb8338ba25a04bdb2de3adcf8d45c7b8eaab0651.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "916338e66e3bbca26f2d59670cc0ed623068c316cefcec91f21124a78802dd7f", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "2f4fa99f13a951bf094007634399536a1c739721629490778fe31da73191ecc5", "generation_seconds_batch": 21.53266403498128, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8982131481170654, "target_z": -0.6679687299267972, "ce": 6.022613525390625, "genuineness": 4.444586753845215, "identity": 0.3315683603286743, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen daar samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/916338e66e3bbca26f2d59670cc0ed623068c316cefcec91f21124a78802dd7f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "918c052a0ab47b3153fcb42388dda121a8c74a0cdcf189035be06754fcea58e8", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "ce76571ec00d63e7fef3df534c0f15720c6efe08961912c33d57f67828e9c735", "generation_seconds_batch": 23.810764549998567, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7278897762298584, "target_z": -0.017700249453121285, "ce": 6.125868320465088, "genuineness": 4.469690799713135, "identity": 0.7537335157394409, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/918c052a0ab47b3153fcb42388dda121a8c74a0cdcf189035be06754fcea58e8.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "9196a1aa22922e61c366f4d7dc5db68f49d729cd85c7be3f55d79fcfbf3c453f", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0ad585a86d5fe28face5dcd2f4b6dd3a2872d716c9da5f483180f299c7798d50", "generation_seconds_batch": 16.244926756015047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.047020435333252, "target_z": 0.23242202919869387, "ce": 5.71682071685791, "genuineness": 4.720732688903809, "identity": 0.7020917534828186, "hypothesis": "J'ai trouvé la 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_CASU_reg/9196a1aa22922e61c366f4d7dc5db68f49d729cd85c7be3f55d79fcfbf3c453f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "91a83642665a1a5168676e1d7b47d12cdb73669a9f59ba3a5764cd82f707d36d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "227077c15d283342ef9baa6f03a6c9cd32de0ec494dbe1e02e784bc39f52e2e6", "generation_seconds_batch": 22.59607640898321, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7369461059570312, "target_z": -1.5781249635032677, "ce": 6.372091293334961, "genuineness": 4.331617832183838, "identity": 0.457396924495697, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/91a83642665a1a5168676e1d7b47d12cdb73669a9f59ba3a5764cd82f707d36d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "922dad8fd0c69b458a8ba79c9c7eca9bbed7187140cab4e837910bf777d25d02", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "2b8b0d01afe18f72f7ab0c70fe8b6d84847bc5b29e5aa33058881f05622720df", "generation_seconds_batch": 21.701490371953696, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6439409255981445, "target_z": -0.08349611459875832, "ce": 5.903473377227783, "genuineness": 4.7081804275512695, "identity": 0.588023841381073, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/922dad8fd0c69b458a8ba79c9c7eca9bbed7187140cab4e837910bf777d25d02.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "92725b4de795464a43b4f332aee286d40b3c7a0145bdf8960e959f00cf507534", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "cc2124fe6b3b5cf6a32641504844eb586c0ac3c926b0fe8a74eb21839a6076e5", "generation_seconds_batch": 23.706792637938634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.773613691329956, "target_z": -0.7656249693427449, "ce": 6.09409761428833, "genuineness": 4.143336772918701, "identity": 0.3593730628490448, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/92725b4de795464a43b4f332aee286d40b3c7a0145bdf8960e959f00cf507534.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "9298ebef22977ff9e3d0b68ac58e343c1b03bb491ef4f7304f43b56a023f5558", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "efc6d67e03f01d84a5619e68c6c1473e319a949b35af32f22dea91d3fea51e46", "generation_seconds_batch": 15.876746883965097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1673479080200195, "target_z": -0.45703130511006573, "ce": 5.855817794799805, "genuineness": 4.7458367347717285, "identity": 0.6987037658691406, "hypothesis": "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_CASU_reg/9298ebef22977ff9e3d0b68ac58e343c1b03bb491ef4f7304f43b56a023f5558.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "92a41c414dd1db69491834ae18ae0bab73edafe510d62241b2e01b444733ad12", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "171df9e2fb5590f0c54f4b35b30a84ce50c490cf720972cb207dfd153b60017c", "generation_seconds_batch": 21.714354204945266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5343828201293945, "target_z": -0.9531248817505874, "ce": 5.982900142669678, "genuineness": 4.582659721374512, "identity": 0.46472394466400146, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/92a41c414dd1db69491834ae18ae0bab73edafe510d62241b2e01b444733ad12.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "935901a30f4857355e19eb4a0014c77dbb039ef362d6c95c0515dfd196d194fe", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "7164e52c53eda5548078367b49b961c7a9e73d102a2498019c1d733adc3ea50c", "generation_seconds_batch": 21.25572887598537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.453214645385742, "target_z": 0.5507813547456216, "ce": 5.871703147888184, "genuineness": 4.3504462242126465, "identity": 0.4681047797203064, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/935901a30f4857355e19eb4a0014c77dbb039ef362d6c95c0515dfd196d194fe.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "93f3261c588fe314d88c5dd73ddb9709f3abac6eb6a32c6f02fefc6cbe22cce7", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "3807815a473d93e8e2d692ac8495254fd7bcff9e3e89063098c6ec20b750d50e", "generation_seconds_batch": 23.75564559700433, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1399359703063965, "target_z": -0.47851572262875885, "ce": 5.9431867599487305, "genuineness": 4.557555198669434, "identity": 0.6378099918365479, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/93f3261c588fe314d88c5dd73ddb9709f3abac6eb6a32c6f02fefc6cbe22cce7.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "940d685122873bb187af683ecd79a1bb182dfb022b033097a3a5338e1c42ef1d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "3ce40b235ead7281a51e4b94474056282994d7a5c556d48c0f487fe541e93f73", "generation_seconds_batch": 19.496425554039888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.252075672149658, "target_z": -0.3906249576637906, "ce": 6.0305562019348145, "genuineness": 4.030367851257324, "identity": 0.2702689468860626, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/940d685122873bb187af683ecd79a1bb182dfb022b033097a3a5338e1c42ef1d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "94154e5d6ce42f06b7ed0fd3045c29a520344204899da985320739b957b2cf16", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "420a5e7fa75d01624726bc867ea205130b819216013f54691c6c227cf8323784", "generation_seconds_batch": 22.718367752968334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.443246364593506, "target_z": 0.5429686014582996, "ce": 5.824047088623047, "genuineness": 4.6705241203308105, "identity": 0.7476417422294617, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/94154e5d6ce42f06b7ed0fd3045c29a520344204899da985320739b957b2cf16.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "949daca123b48cc1407780e1b95616914904f49a254f91faab28727989b88ac0", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "98e1dfcf58aa4fa142f32acb0d97a16323f7d6258c927ae349f3a8bc309945a3", "generation_seconds_batch": 25.063222842058167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.145632266998291, "target_z": -1.257812543066144, "ce": 6.09409761428833, "genuineness": 4.093128204345703, "identity": 0.5078537464141846, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/949daca123b48cc1407780e1b95616914904f49a254f91faab28727989b88ac0.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "94a1c352b4266c341f6d9e714f10e39933945b4aa2e8081f45c8f75d7d8f204d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "d3c68256c36891479bf94b3eed1e15e33da2948f76203f785c1334c722b0d795", "generation_seconds_batch": 18.051342872087844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4745750427246094, "target_z": -0.9999999065683653, "ce": 6.022613525390625, "genuineness": 4.494794845581055, "identity": 0.5101746320724487, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/94a1c352b4266c341f6d9e714f10e39933945b4aa2e8081f45c8f75d7d8f204d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "95a17f0de0539cf7bc2ed464f98e4f6e8f98dc87226c91a4badb67fa2127858d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e7ba7956cee2fd388893155f1ac32f4c41f60320076330d93ebce7618c687262", "generation_seconds_batch": 23.90744831704069, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5212104320526123, "target_z": -0.17968753284705904, "ce": 6.308549880981445, "genuineness": 4.457139015197754, "identity": 0.20670071244239807, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/95a17f0de0539cf7bc2ed464f98e4f6e8f98dc87226c91a4badb67fa2127858d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "96253fa15245b5b7fe1db301aa422dc4c1e56d3b4822715006c83f85709a4663", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7342cf175bfd73279b5565c6e73687a9ce45eaf88f12a739707533b637ba1683", "generation_seconds_batch": 22.555652193026617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.458198547363281, "target_z": 0.5546875445260133, "ce": 5.895530700683594, "genuineness": 4.833701133728027, "identity": 0.6187759637832642, "hypothesis": "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_CASU_reg/96253fa15245b5b7fe1db301aa422dc4c1e56d3b4822715006c83f85709a4663.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "962b4406aa5df9167e27505bbb15c04b4b7efc51abd3ab3947539de48c76626e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "89efad6f36992eb57c0b8f0c4bb56ae899517ac08e760211758476ffe6421b1a", "generation_seconds_batch": 26.318961848970503, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3442792892456055, "target_z": -0.31835932554692775, "ce": 5.927301406860352, "genuineness": 4.457139015197754, "identity": 0.5421397089958191, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/962b4406aa5df9167e27505bbb15c04b4b7efc51abd3ab3947539de48c76626e.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": "963fe249fb43e9f8f0cc7947c13de698b57cbdb3c7bd770156b3af36c1af937e", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9164a767a33121781dc1514781e3be718f48864c2b99f3816f372c47d708e8ce", "generation_seconds_batch": 25.09243169601541, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.493086338043213, "target_z": 0.582031246715294, "ce": 5.903473377227783, "genuineness": 4.582659721374512, "identity": 0.6519557237625122, "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_CASU_reg/963fe249fb43e9f8f0cc7947c13de698b57cbdb3c7bd770156b3af36c1af937e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "969e7ee831cc4b3f440dd3831c088bce045b73bdd7e166c497bcbe710bb359f6", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d99103205c58b2c73a45794c8fc136447c64702a07f8a504abd4b9b6514748f6", "generation_seconds_batch": 21.284632114926353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6676149368286133, "target_z": -0.06494133941535903, "ce": 6.09409761428833, "genuineness": 4.017816066741943, "identity": 0.27817827463150024, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/969e7ee831cc4b3f440dd3831c088bce045b73bdd7e166c497bcbe710bb359f6.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "96ac2dca319df40e455042948ec9821ae1df8569c8f631b5a5f547fde7304714", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "50f210bce18ec7fa359a341dc6332d5184c62a6e70bcd927f7f12a9d09065759", "generation_seconds_batch": 23.627107370994054, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.546130418777466, "target_z": -0.16015621021856183, "ce": 5.792276382446289, "genuineness": 4.137060642242432, "identity": 0.3427632451057434, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/96ac2dca319df40e455042948ec9821ae1df8569c8f631b5a5f547fde7304714.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "96b7f7d748065b0dffd2e7926c53317a19d41023fcd2d9e50299c65094259400", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "efac527355bc96841707b99fce414caa67f86004ca4bf7e0209a4162d81d1f4a", "generation_seconds_batch": 22.56079834396951, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.134239673614502, "target_z": 0.3007810978086264, "ce": 6.038498878479004, "genuineness": 4.344170093536377, "identity": 0.6078277826309204, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/96b7f7d748065b0dffd2e7926c53317a19d41023fcd2d9e50299c65094259400.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "96c66dabbbbc8b623fca4caca515f402f1f796ff23eb3fb2d888dc02a53ea252", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "a026abdedef5f83e219b35c18890fdde5b268803cee7a753855021e284b10004", "generation_seconds_batch": 27.112057235906832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4763545989990234, "target_z": -0.2148438014603925, "ce": 5.824047088623047, "genuineness": 4.482243061065674, "identity": 0.8350908756256104, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/96c66dabbbbc8b623fca4caca515f402f1f796ff23eb3fb2d888dc02a53ea252.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "96e6943634e0c29ea4e230ee58e4d81f1b387c8c544cd852b934ac9dec6a59e8", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "8c23bd92ccecc1eb9c5b3feb044a97aa0a751dbaf5100f3fb2b57a758e0a12b9", "generation_seconds_batch": 22.25931234296877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.722349166870117, "target_z": 0.7617185926990839, "ce": 5.895530700683594, "genuineness": 4.494794845581055, "identity": 0.6454212665557861, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/96e6943634e0c29ea4e230ee58e4d81f1b387c8c544cd852b934ac9dec6a59e8.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": "97099c7558b333fb267d8522364f538f7c9ccc7f4343db348aa6299ed87423bf", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a15017d40315c25a567063612e848b9da600ec86f6aafe3337b1fbb8722995f", "generation_seconds_batch": 22.77398750500288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.356027126312256, "target_z": 0.4746095328483671, "ce": 5.879645347595215, "genuineness": 4.406930446624756, "identity": 0.6133791208267212, "hypothesis": "La route tourna à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.05, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/97099c7558b333fb267d8522364f538f7c9ccc7f4343db348aa6299ed87423bf.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "972cda8a7cbb589613b0c6c50042e12c3b7fca14ad43d9d7e0699470aeae417b", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "f1e8243964a57a4ba21249231e2e98d11bf8f20b45e5676095f192558fca5893", "generation_seconds_batch": 16.456124094082043, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7967537641525269, "target_z": -1.5312500321171243, "ce": 5.9431867599487305, "genuineness": 4.231201171875, "identity": 0.7989946007728577, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/972cda8a7cbb589613b0c6c50042e12c3b7fca14ad43d9d7e0699470aeae417b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "97816d11cccc6be50c8d4b057bc59c8cbe69e833db258f4c24a085a39a411097", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e39f7f2dd0fd7261026dd9156eb5523ed30918881029ebd53dd7d153962a6f79", "generation_seconds_batch": 20.998760886024684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.858875036239624, "target_z": 0.08496093576640522, "ce": 6.260893821716309, "genuineness": 4.595211505889893, "identity": 0.4301501512527466, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/97816d11cccc6be50c8d4b057bc59c8cbe69e833db258f4c24a085a39a411097.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "97942eea869c1d5a02fb59b6f2992b525bd78346af9b48e318fd2294f2cd91ab", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "b2e327796f21147179cffdd9719efee991f99d7ef4149f35f2be40abb34620a7", "generation_seconds_batch": 24.977417498943396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4845428466796875, "target_z": -0.9921875270075818, "ce": 6.078212261199951, "genuineness": 4.394378662109375, "identity": 0.7359840869903564, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/97942eea869c1d5a02fb59b6f2992b525bd78346af9b48e318fd2294f2cd91ab.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "985103605138d941699811d067e57e4d8b109ee5e435e64f06c1a21ec573205a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "0de1106716393e4df02601acad672eca95e82f65c3a4c33d2f4b5813eba80d15", "generation_seconds_batch": 25.129749748972245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4913065433502197, "target_z": -0.20312504525594802, "ce": 6.078212261199951, "genuineness": 4.582659721374512, "identity": 0.7283385396003723, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/985103605138d941699811d067e57e4d8b109ee5e435e64f06c1a21ec573205a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "9904ef528026f38b586bd83e1087cc4edca6c8953e9d8acd1a21f6af003c82a5", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "a70deca277541c308c0cb3d7e9b60ddff181df36a18cce13e72a2e3d1bd346dc", "generation_seconds_batch": 22.663959964062087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7118477821350098, "target_z": -0.03027334452457453, "ce": 5.831989288330078, "genuineness": 4.193545341491699, "identity": 0.5589125156402588, "hypothesis": "Ik 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_CASU_reg/9904ef528026f38b586bd83e1087cc4edca6c8953e9d8acd1a21f6af003c82a5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "9909836bf09906275802886a0590af6e877b26ddbae365ff46e36d8e3ad923ab", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "9c5c73c8e9005e934b044bd1d9ad6d98a8057ec3b42c176ee2ab9a773b32943d", "generation_seconds_batch": 22.661100882920437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.383439064025879, "target_z": 0.49609395036706017, "ce": 5.4666266441345215, "genuineness": 4.645420074462891, "identity": 0.7633141875267029, "hypothesis": "Oh. La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/9909836bf09906275802886a0590af6e877b26ddbae365ff46e36d8e3ad923ab.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "992aed85da20e41b4089cd7e1f53121a0fb91dbed50bd29ae28a062de0995b62", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "cb9f619b28bbe918d44c60863a749a34088c99b5e92a326f3d897d522e3d7330", "generation_seconds_batch": 22.892663629027084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.032780647277832, "target_z": -0.5624999240867968, "ce": 6.078212261199951, "genuineness": 4.582659721374512, "identity": 0.6622294187545776, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/992aed85da20e41b4089cd7e1f53121a0fb91dbed50bd29ae28a062de0995b62.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "992fa44fb6aff5af2add5ace74e4a1f92be8bf6254a8ef1c79070e5243bcccda", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "c54e71ed96d87e85a6ce765eb3998907be57edaebdb89c32094a1c14876a0438", "generation_seconds_batch": 22.557140112970956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8688430786132812, "target_z": 0.09277350219045796, "ce": 6.133810997009277, "genuineness": 4.281409740447998, "identity": 0.5742954015731812, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/992fa44fb6aff5af2add5ace74e4a1f92be8bf6254a8ef1c79070e5243bcccda.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "99be1d0cadbd9b1ec46ec538a9c4021b7e066f729b3b7538b42bfd3a9559d607", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a51a9d8b5b067948ce5753ea88767c30819b8db658f86b7ae0a82252d50e6892", "generation_seconds_batch": 17.66382387606427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.523702383041382, "target_z": -0.17773443795686317, "ce": 5.95112943649292, "genuineness": 4.268857479095459, "identity": 0.6552952527999878, "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_CASU_reg/99be1d0cadbd9b1ec46ec538a9c4021b7e066f729b3b7538b42bfd3a9559d607.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "81be090a8fac93cd5c880cc6a237651020243fa60be9d6f9b259b15466aaea90", "generation_seconds_batch": 16.802694446058013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8932290077209473, "target_z": -0.6718751065704582, "ce": 6.395919322967529, "genuineness": 4.507347106933594, "identity": 0.603705644607544, "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_CASU_reg/99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "9ad1d3eb16b6cf81e147580447085c81da92fa85095421fdde4db2635d3f215b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d7037238f09dd5a741e3d71aca720e429cd349366baa8273ebb732427ff8a15b", "generation_seconds_batch": 21.284632114926353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.187283992767334, "target_z": -0.4414061722619603, "ce": 6.157639026641846, "genuineness": 4.319066047668457, "identity": 0.3533216714859009, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/9ad1d3eb16b6cf81e147580447085c81da92fa85095421fdde4db2635d3f215b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "9b39459942d902fde90444d77819769a22d054855e6cab7dcad4d5a91572a998", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "eac6c471e1ad1fe5f4c07176066dbfa75f12dfb971ab73762a68b966d8f4a3b8", "generation_seconds_batch": 23.129906447022222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.365994930267334, "target_z": 0.48242191240915056, "ce": 5.6334228515625, "genuineness": 4.306513786315918, "identity": 0.5083770751953125, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen 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_CASU_reg/9b39459942d902fde90444d77819769a22d054855e6cab7dcad4d5a91572a998.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "9b79a5e7af256759ce2168023d4ce811f8bad1a5ab2edd0d66bab149c945c1ad", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "218e898837489f03bd1afdbd4a3497c5fc5496805df283505e48f332ee5de9a5", "generation_seconds_batch": 19.21011547197122, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9018619060516357, "target_z": 0.11865238321209177, "ce": 6.300607204437256, "genuineness": 4.532451152801514, "identity": 0.3350183665752411, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/9b79a5e7af256759ce2168023d4ce811f8bad1a5ab2edd0d66bab149c945c1ad.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "9bf67f412f764746db32d361e90580895c1e2b81eca94942e0a2b841a9638959", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1cf333a8ed0eb0597a8ccd9e1719ef0591f3039d0bcaff183d8808af14df5610", "generation_seconds_batch": 23.46319490496535, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7336525917053223, "target_z": -0.01318357737209139, "ce": 5.990842819213867, "genuineness": 4.570107460021973, "identity": 0.6978880167007446, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/9bf67f412f764746db32d361e90580895c1e2b81eca94942e0a2b841a9638959.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "3bf46b8b4dece2d35a9de740305ee92edebcd7432401018a853215688aaac915", "generation_seconds_batch": 16.888734317035414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2321395874023438, "target_z": -0.40625009051189603, "ce": 6.117925643920898, "genuineness": 4.356721878051758, "identity": 0.6703252792358398, "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_CASU_reg/9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "9c0f58b5d4a436cc658a4c1ceb31426424c979fd8a6eb26d3a648bab7778f7ac", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "abf77d8a358972d4001dd743aecc6a5c052ea616de70bb1fb903550e2432cd67", "generation_seconds_batch": 21.66386050102301, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.728045701980591, "target_z": -0.017578040875032105, "ce": 5.80021858215332, "genuineness": 4.808597087860107, "identity": 0.5829808115959167, "hypothesis": "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_CASU_reg/9c0f58b5d4a436cc658a4c1ceb31426424c979fd8a6eb26d3a648bab7778f7ac.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "9c2f183b3a81efdb512c4960083a903d682c1e778b0920d3903f948ff09b7aa4", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "e6142b53f26c1793e94cc69735d746840d46cf8fffc5f05061ecfd406c56c7ba", "generation_seconds_batch": 22.68062383902725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3948311805725098, "target_z": -1.0625000642342488, "ce": 5.593709468841553, "genuineness": 4.168440818786621, "identity": 0.6627706289291382, "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_CASU_reg/9c2f183b3a81efdb512c4960083a903d682c1e778b0920d3903f948ff09b7aa4.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe35099bffd03034017b4c85e4025d84cf5c1e2acc84e2de8bf8281212408dfb", "generation_seconds_batch": 16.737210438936017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0060808658599854, "target_z": -1.3671875386865362, "ce": 6.038498878479004, "genuineness": 4.306513786315918, "identity": 0.7232190370559692, "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_CASU_reg/9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "9d30390226e453fdcb4c32d12be47c69ff87310b979194b6111007327774181c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "7cf6c57ee4bf103f30aa1e96b9b321e8b81496e5c11da40a2c800c008d1d6b8d", "generation_seconds_batch": 23.910638179047965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1025562286376953, "target_z": -0.5078125197082354, "ce": 6.125868320465088, "genuineness": 4.645420074462891, "identity": 0.6409780979156494, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/9d30390226e453fdcb4c32d12be47c69ff87310b979194b6111007327774181c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "9d31261f8b85167c48470cc7314f7148ed75e4212c22396f3f2afce216acea5a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e8a2d763c76670c45bca1ff32dc2449fe7ee2be238f96d4514bf01ac21f6e17", "generation_seconds_batch": 21.483801538008265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8665294647216797, "target_z": -1.4765625343069284, "ce": 6.475345611572266, "genuineness": 4.381826400756836, "identity": 0.6081990003585815, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/9d31261f8b85167c48470cc7314f7148ed75e4212c22396f3f2afce216acea5a.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": "9db31244b909f0c1d18f836c3d1ad6dfd42ce82c8069272b4fbf03f087907742", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "048e017ae953fc61e880033f6578626faf4dcbaf3807419389edc2bbfeecded8", "generation_seconds_batch": 17.190951670054346, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.142428159713745, "target_z": -0.4765624408752937, "ce": 6.086154937744141, "genuineness": 4.482243061065674, "identity": 0.6779091358184814, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, ließ sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/9db31244b909f0c1d18f836c3d1ad6dfd42ce82c8069272b4fbf03f087907742.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "9eca9a74bc6707e190e1ea5e581ae4d0c91b3ec72d36e8df3fdce7ff5c4c9753", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "c2d85ef74c21e13a8748616e97d2a1f7cd749dbd8a30b334b556ba0088655c47", "generation_seconds_batch": 22.683037291979417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.29871129989624, "target_z": 0.429687602920785, "ce": 6.181467056274414, "genuineness": 4.482243061065674, "identity": 0.6144617199897766, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/9eca9a74bc6707e190e1ea5e581ae4d0c91b3ec72d36e8df3fdce7ff5c4c9753.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "9f1d2dd31feb45e30103e839e10691b172763780fe5e2c5ae2ebb5a7041ee66d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "51bd1e4a39bfa5e7e020e99da98ac079ca16be28e86a386769d5d7de139ac7c0", "generation_seconds_batch": 24.726403250941075, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0626845359802246, "target_z": -0.5390624116779079, "ce": 6.046441555023193, "genuineness": 4.68307638168335, "identity": 0.7623698711395264, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/9f1d2dd31feb45e30103e839e10691b172763780fe5e2c5ae2ebb5a7041ee66d.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": "9f936b9062cba49580a51df517af2bba701ffa4b0c9f8c8e841ca9bbf9b3f125", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2469825827aa134c0f4fa8ea62d715ba6bce1e4a7ba9eb9713b90527e79ecacf", "generation_seconds_batch": 24.549607152934186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4975366592407227, "target_z": -0.1982421211671891, "ce": 5.363371849060059, "genuineness": 4.557555198669434, "identity": 0.7259182929992676, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zu dem kleinen Holstor. Ich warte auf der anderen Seite.", "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_CASU_reg/9f936b9062cba49580a51df517af2bba701ffa4b0c9f8c8e841ca9bbf9b3f125.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "a145b23574e1cbed05ed3c7a02c4d6be04895d8c37922fdaae7afd65a2dace5c", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "e917de4921b6af7b1d3d857b110a9221d56967d89defe157886749d30373cb1c", "generation_seconds_batch": 16.16778785700444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0975723266601562, "target_z": -0.5117187094886272, "ce": 6.141753673553467, "genuineness": 4.482243061065674, "identity": 0.2791904807090759, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a145b23574e1cbed05ed3c7a02c4d6be04895d8c37922fdaae7afd65a2dace5c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "a1dd0bf8ee29f4ca26ef67e4d8207f492b2641835a6901686cd254906b28bcde", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "21e8349432bde09fe27c38cbd4c140b8c82f2a3ae8eea89114892b3eee4e0f4d", "generation_seconds_batch": 22.192778203985654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4800925254821777, "target_z": -0.2119141591250987, "ce": 6.086154937744141, "genuineness": 4.457139015197754, "identity": 0.6638757586479187, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a1dd0bf8ee29f4ca26ef67e4d8207f492b2641835a6901686cd254906b28bcde.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "a1f76d979eeee5de2e4831c708ca216b22419c2c90e7743c6591a95e92056d79", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "7392875006d0cc899329c419948d29474c4dc473f28fa3d7e3d3692e44b18909", "generation_seconds_batch": 16.514187742955983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0527164936065674, "target_z": -0.5468749781019606, "ce": 6.133810997009277, "genuineness": 4.645420074462891, "identity": 0.671159565448761, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a1f76d979eeee5de2e4831c708ca216b22419c2c90e7743c6591a95e92056d79.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "a23f07810a61d7f1e4d5b7426c1ad8316cd7604a5f2250837b61af1b22020df0", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "18cca22512db0ad82721c92d47a66247bedf5309e75cf56588b5fdfdbc4962e1", "generation_seconds_batch": 21.25572887598537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3716912269592285, "target_z": -0.2968749080282347, "ce": 5.77639102935791, "genuineness": 4.344170093536377, "identity": 0.13239938020706177, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a23f07810a61d7f1e4d5b7426c1ad8316cd7604a5f2250837b61af1b22020df0.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "a25b412ebaace354c4f84c51ff2091173546b554350252093abaee140e21aa7d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "2e31c57112451f527fb2038e612ac0e6c2f1a39e9f887d9d6b619da311bb04e5", "generation_seconds_batch": 23.138383396086283, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.672910213470459, "target_z": -0.06079110620532741, "ce": 6.078212261199951, "genuineness": 4.457139015197754, "identity": 0.8187221884727478, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a25b412ebaace354c4f84c51ff2091173546b554350252093abaee140e21aa7d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "a2866db479d3561eebc4d5adfaee4da1cec69dd7cb3cad550a9d045009933d5f", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "55562ebfe6a44922a484916a3c80fa574540146df53e592c01d84829341e6bac", "generation_seconds_batch": 16.16778785700444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2353439331054688, "target_z": -1.1875000058394771, "ce": 6.23706579208374, "genuineness": 4.13078498840332, "identity": 0.13567900657653809, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a2866db479d3561eebc4d5adfaee4da1cec69dd7cb3cad550a9d045009933d5f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "a2b92b789bc2eae435b31eef868f1d427c4ad54bceebec4adfc978b762e00540", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "17410f2c66e0e28ea590368f0a41b56a456f119d93aa08fb7c9cf8dc6d3969e5", "generation_seconds_batch": 22.84490949800238, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.559302568435669, "target_z": -0.9335937459853595, "ce": 6.205295085906982, "genuineness": 4.632867813110352, "identity": 0.5715445280075073, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a2b92b789bc2eae435b31eef868f1d427c4ad54bceebec4adfc978b762e00540.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "a371385aafe8710cfbf90296c0d93a66c3057e98badab863e1fb292f163903dc", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "7f7e125d88d90547b2b7b875f6e07ac79b24582d27f564fc79424a96cc3275be", "generation_seconds_batch": 23.535603750962764, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.763645648956299, "target_z": -0.7734375357667976, "ce": 5.8875885009765625, "genuineness": 4.632867813110352, "identity": 0.5902332067489624, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a371385aafe8710cfbf90296c0d93a66c3057e98badab863e1fb292f163903dc.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "a38dc7080befdcb46d81bbe794885d41b6c5b51525606a1b5b50c4d7f05be330", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "031ff9481229251eb6ebb28ec36fa025eb592eb5c23e07e29268e954409ee408", "generation_seconds_batch": 22.471012047026306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.863325357437134, "target_z": -0.695312432116078, "ce": 5.974957466125488, "genuineness": 4.65797233581543, "identity": 0.7031019926071167, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a38dc7080befdcb46d81bbe794885d41b6c5b51525606a1b5b50c4d7f05be330.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "a3e6cbc3ccb3aa31ada93938e631c6fb786c872ce3679d4d51c1b6427f845485", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "277e60832919d68eab89a9009cdfe024f88b0f4262b24a9c930141aa82998069", "generation_seconds_batch": 17.66382387606427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.414767265319824, "target_z": -1.0468749313861434, "ce": 5.8875885009765625, "genuineness": 4.268857479095459, "identity": 0.5196071863174438, "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_CASU_reg/a3e6cbc3ccb3aa31ada93938e631c6fb786c872ce3679d4d51c1b6427f845485.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "a43b10ffce6c391c182776b4d65efae45b70f3dc69523a2d67f40afc0873cfd1", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "d68e334dcf098c5269001595bfc193eb09eef3abb53b5dc0eb57021ac4821a52", "generation_seconds_batch": 25.470492660999298, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7835817337036133, "target_z": -0.7578124029186921, "ce": 6.086154937744141, "genuineness": 4.457139015197754, "identity": 0.6433712244033813, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a43b10ffce6c391c182776b4d65efae45b70f3dc69523a2d67f40afc0873cfd1.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "a4ae293d15b713ce3f78d4e73843c3b4170ca44f777a4b63fd67bc2507ff8667", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b5876533894480c8d1c1b1168b86c05f611ca8138a2fac0b7a598db2af36f876", "generation_seconds_batch": 19.071058742003515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.496290683746338, "target_z": -0.19921866861228701, "ce": 5.665193557739258, "genuineness": 4.482243061065674, "identity": 0.6968770027160645, "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_CASU_reg/a4ae293d15b713ce3f78d4e73843c3b4170ca44f777a4b63fd67bc2507ff8667.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": "a4b43645087b1b8498cad304997bc7c91feb8f79ce6f68e36dd612233f564cb7", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I wrote the number on the envelope.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "62ec62d44e9d9cc7521b6506fb9d87aa264f142aa93ed55623eb052ea923b51b", "generation_seconds_batch": 17.15580042800866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0178287029266357, "target_z": -0.5742186802912413, "ce": 6.1894097328186035, "genuineness": 4.061748504638672, "identity": 0.5647613406181335, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a4b43645087b1b8498cad304997bc7c91feb8f79ce6f68e36dd612233f564cb7.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "a52cbcc38fba0d36c55b20dc5e3487ddac24566dff47fc1f7801e094c62f02fd", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8cae72bf8f24858300b1217e20d71a4c1aa9e76c3beff036d3c65fc3e7908e36", "generation_seconds_batch": 23.179981578025036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9710144996643066, "target_z": 0.17285151386810424, "ce": 6.0305562019348145, "genuineness": 4.243753433227539, "identity": 0.3743138313293457, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a52cbcc38fba0d36c55b20dc5e3487ddac24566dff47fc1f7801e094c62f02fd.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "a56fc4ac9879c7edde48c49df21601b61ee33fb6424bd9ef134dbd983910bcb5", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "b6903655cd1cedcb123cec49c1404f28eddcbdc475db5c7ae63168462cb342b1", "generation_seconds_batch": 22.044438773067668, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9672765731811523, "target_z": 0.16992187153281044, "ce": 6.181467056274414, "genuineness": 4.080576419830322, "identity": 0.4744853973388672, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a56fc4ac9879c7edde48c49df21601b61ee33fb6424bd9ef134dbd983910bcb5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "a5d621a17dd4ce51bd9955e3dc6f81432c468dee06f2b6fc5ca49163136f3890", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d08d671ead399872c19195c134e6684316ab340f5c624ff3f503455648c4f3b2", "generation_seconds_batch": 22.74679440294858, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6470558643341064, "target_z": -0.08105474598601348, "ce": 6.133810997009277, "genuineness": 4.68307638168335, "identity": 0.8166295289993286, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a5d621a17dd4ce51bd9955e3dc6f81432c468dee06f2b6fc5ca49163136f3890.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": "a5fc2c30e7c079c1824d137daca28d0275a8a4270e73b2b79aeaa98833227fd7", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4caa97b02e9d79e0d1248d3c0bb770bd2977051e18051a4450014c2097bec1df", "generation_seconds_batch": 22.663453585933894, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3866429328918457, "target_z": -0.2851563386870594, "ce": 5.732706069946289, "genuineness": 4.582659721374512, "identity": 0.7759109735488892, "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_CASU_reg/a5fc2c30e7c079c1824d137daca28d0275a8a4270e73b2b79aeaa98833227fd7.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "a7c34915eb1d536ac144089f445a9e2af6e037c3d94fd8e10bf95e0bc2d5297d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "d30312451706c1f745e62126eebe8f8eb19db7f9f9a049231826014298c773f2", "generation_seconds_batch": 17.096238186000846, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.061972141265869, "target_z": 0.2441405985398691, "ce": 5.760505676269531, "genuineness": 4.65797233581543, "identity": 0.7514933347702026, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a7c34915eb1d536ac144089f445a9e2af6e037c3d94fd8e10bf95e0bc2d5297d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "a7c80b1eea387940e0cda970a3be04ae7da61271504559285a77b876a0621ae3", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a3c4cf4e2a5465d1c10166e5c6c4f9b6d6edabd839d08ca238e5324aeaff2d86", "generation_seconds_batch": 16.865278732031584, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.29871129989624, "target_z": 0.429687602920785, "ce": 5.784333229064941, "genuineness": 4.595211505889893, "identity": 0.7651926279067993, "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_CASU_reg/a7c80b1eea387940e0cda970a3be04ae7da61271504559285a77b876a0621ae3.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "a7ca5794c65ad8d92ba803874b6fc60174fae7db5fb37246bc1c936100763997", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "152f52536147b20bd685bbafbbd26c833d2bed6b9542f64ea9d8c6571159c007", "generation_seconds_batch": 19.89327829494141, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6744678020477295, "target_z": -0.05957032846732037, "ce": 5.764476776123047, "genuineness": 4.570107460021973, "identity": 0.7622538805007935, "hypothesis": "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_CASU_reg/a7ca5794c65ad8d92ba803874b6fc60174fae7db5fb37246bc1c936100763997.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "a7dd4516885bde697659001c274ed8aed5255c51acae8673c0f58dc1eb3eb633", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7e4a07640f44ccb9a1593a9535a0ed490bffdbe01e8f1836f099700d995950ec", "generation_seconds_batch": 16.678186065983027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.863325357437134, "target_z": -0.695312432116078, "ce": 5.967014789581299, "genuineness": 4.457139015197754, "identity": 0.587585985660553, "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_CASU_reg/a7dd4516885bde697659001c274ed8aed5255c51acae8673c0f58dc1eb3eb633.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "a7f0fb30cc654745e2ce0d31b0035aeee25a74cc932de9cbe09df0f748c1fc69", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "29d9ea3e4321990fec63c1713bacda248bec9573ac7ba4dd8ebf618a17987e1e", "generation_seconds_batch": 22.65963611099869, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5324244499206543, "target_z": -0.17089841897790836, "ce": 5.959072113037109, "genuineness": 4.055472373962402, "identity": 0.5199853181838989, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a7f0fb30cc654745e2ce0d31b0035aeee25a74cc932de9cbe09df0f748c1fc69.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "b6eb7418803b4b7b2cbbb30eb3cc5c7ebfd73473e7e840754edd12137761a52d", "generation_seconds_batch": 17.05230904999189, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.404799222946167, "target_z": -1.054687497810196, "ce": 6.181467056274414, "genuineness": 4.469690799713135, "identity": 0.7011718153953552, "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_CASU_reg/a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "a91316df42feb2d1addcd28e1ce96131960bf4c5a8bee8f8c54e9bc340cec320", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "56332fb7f73002802c9417a883534994cad4e0c03f39ccf8f0099ac66f00fc3b", "generation_seconds_batch": 22.606808132957667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.607807159423828, "target_z": -0.11181636423313697, "ce": 5.895530700683594, "genuineness": 4.582659721374512, "identity": 0.8103696703910828, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a91316df42feb2d1addcd28e1ce96131960bf4c5a8bee8f8c54e9bc340cec320.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": "a944a5343be5389d4daac3b9ce88826c839af9c1dab36ffc0b06a553868289eb", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "fdd0ac9b33d99e329c37dfd44c36cad6a79904e31a75fa29c877d32d6bb9aa5b", "generation_seconds_batch": 22.728559091105126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2794876098632812, "target_z": -0.36914054014509745, "ce": 5.752562999725342, "genuineness": 4.582659721374512, "identity": 0.6761046648025513, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue, elle était ici, près de la grande bibliothèque nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a944a5343be5389d4daac3b9ce88826c839af9c1dab36ffc0b06a553868289eb.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "cf368978d1955bb7c4b36dc325db3f93ded6e4894fc7110f4b3a7b76d7ed9007", "generation_seconds_batch": 18.585101302014664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3748953342437744, "target_z": -1.0781250102190851, "ce": 6.125868320465088, "genuineness": 4.243753433227539, "identity": 0.71775221824646, "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_CASU_reg/a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "a9dc438887e4c6e904b1346a1edaaaf321628083bd977e59c14834f9671d9a63", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5a802bca63cd2821f18ab9f88846486280f06c34f32d65c9dc9022074757426c", "generation_seconds_batch": 22.922871061018668, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7028141021728516, "target_z": -0.03735359379643844, "ce": 6.213237762451172, "genuineness": 4.921566009521484, "identity": 0.818345308303833, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/a9dc438887e4c6e904b1346a1edaaaf321628083bd977e59c14834f9671d9a63.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "aa0c2bc6190f67c93f757d4bea781e38014b3e362a5a6f9dedce2a9fa45308f1", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "86264ec8df65c72daef2cbe53f6c088fb52301a9376721d701d740ad98b0ded7", "generation_seconds_batch": 17.533436162979342, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0769243240356445, "target_z": 0.25585954160758284, "ce": 6.2767791748046875, "genuineness": 4.6705241203308105, "identity": 0.3332688808441162, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/aa0c2bc6190f67c93f757d4bea781e38014b3e362a5a6f9dedce2a9fa45308f1.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "aa6ab5e4e87499da85f8ab95f6c2038f236e69363c5511a2c64de92fb1a30267", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "1aa1b055608709e3c5b7202a7da368e7f4368c03318eb1f6c667353610093a19", "generation_seconds_batch": 24.55935576907359, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1591596603393555, "target_z": 0.32031242043712366, "ce": 5.871703147888184, "genuineness": 4.519899368286133, "identity": 0.5410721302032471, "hypothesis": "Me encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/aa6ab5e4e87499da85f8ab95f6c2038f236e69363c5511a2c64de92fb1a30267.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "aabb42dda414efe10323e0c9e1f205640ae39971d18597d07a60aa7f03ca23b4", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d652ffecb91ac0025d140ee742279143ed9ff9ddf4530c0fcdf562a3974ea984", "generation_seconds_batch": 24.327107973047532, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4327449798583984, "target_z": -0.2490233357653588, "ce": 5.935244083404541, "genuineness": 4.570107460021973, "identity": 0.5524536371231079, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/aabb42dda414efe10323e0c9e1f205640ae39971d18597d07a60aa7f03ca23b4.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "ab40d712c1c77ea08d11dcc32601144df03882aab235376083f2ce3ff58f690b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c0daad42b7a4ba7b2dd5b40a76ba2f546f24082eee534c6613223a02747e98f2", "generation_seconds_batch": 22.922871061018668, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2221717834472656, "target_z": -0.41406247007267954, "ce": 5.95112943649292, "genuineness": 4.570107460021973, "identity": 0.6428934335708618, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ab40d712c1c77ea08d11dcc32601144df03882aab235376083f2ce3ff58f690b.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "f79abdd01ba6806340451fba2b3eb1b1a850724509cadee8113e65a4d9280277", "generation_seconds_batch": 17.376892061904073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.414767265319824, "target_z": -1.0468749313861434, "ce": 6.41180419921875, "genuineness": 4.306513786315918, "identity": 0.34396126866340637, "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_CASU_reg/ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "ab7e4a12e70a4798f09672f8f5df4af4aa6ce00ff34ab0788f3bcf47e5fb4178", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5bf2df6d84c2a44029b380596d4185abe4ebe7dc6dcffe5109d4bbc83acb6796", "generation_seconds_batch": 23.26663729501888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.438262462615967, "target_z": 0.5390624116779079, "ce": 5.982900142669678, "genuineness": 4.3127899169921875, "identity": 0.5666905641555786, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ab7e4a12e70a4798f09672f8f5df4af4aa6ce00ff34ab0788f3bcf47e5fb4178.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "abb0648822d09e06653ed44b00261fe5ec4b74e1d4e26f1a7374ae566dfc86bf", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "15c3cc435a3eabc0481e2a581b1032b7100ef9efc8bcadaac6a61e393c6faab7", "generation_seconds_batch": 16.061606397968717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.071939945220947, "target_z": 0.2519529781006526, "ce": 5.708878040313721, "genuineness": 4.7834930419921875, "identity": 0.7944844961166382, "hypothesis": "La porte est encore ouverte J'ai posé ton manteau sur la chaise dis pas quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/abb0648822d09e06653ed44b00261fe5ec4b74e1d4e26f1a7374ae566dfc86bf.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "abe8d35b75dffed492800c1407b555552284cf27cb14bbc11907c40ab117544c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "fca29f877bae6c39f57c9ee6bb563e78b7375919930bffcab2f9f79d54504618", "generation_seconds_batch": 18.85998826404102, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.373470783233643, "target_z": 0.4882811970797382, "ce": 6.1020402908325195, "genuineness": 4.557555198669434, "identity": 0.7290186882019043, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/abe8d35b75dffed492800c1407b555552284cf27cb14bbc11907c40ab117544c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "abf28342474d03f9f665fecd84866e658a6f865b428bb25d77a5d0ac41ce5989", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "15b162935e8a872253f1a098cbdeea1fd402aa708e8bf43422de1dfbd41d21c3", "generation_seconds_batch": 22.64155014802236, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.206507682800293, "target_z": 0.35742197080392224, "ce": 5.895530700683594, "genuineness": 4.319066047668457, "identity": 0.5344651937484741, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/abf28342474d03f9f665fecd84866e658a6f865b428bb25d77a5d0ac41ce5989.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "ac0863d33902d5a7c7c68ee25e4910ca8acd59f02e907736df4fcccb796c2eda", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "aa4619c07750d686976cc15fd292cacc2f3f82a487dd13b0741f980e11fd1a49", "generation_seconds_batch": 17.434772856067866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4314987659454346, "target_z": -0.250000070073726, "ce": 5.903473377227783, "genuineness": 4.3881025314331055, "identity": 0.7442024946212769, "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_CASU_reg/ac0863d33902d5a7c7c68ee25e4910ca8acd59f02e907736df4fcccb796c2eda.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "ac2812419b3152c2a036b9cd2c4931d75a4df8d906df202606d82f15e3dd64c2", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "82e614c3d1788bb5c619c07e9798e6fe727f59e68ea97e3b2b224f84c2f8d9fe", "generation_seconds_batch": 22.65963611099869, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3118834495544434, "target_z": -0.34374993284601263, "ce": 5.998785495758057, "genuineness": 4.2751336097717285, "identity": 0.341708242893219, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you're ready.", "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_CASU_reg/ac2812419b3152c2a036b9cd2c4931d75a4df8d906df202606d82f15e3dd64c2.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "ac4c4808c198fe50a302a413f18c87b81aff970490732d21f4c3bbd424f9a030", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "4d5642283e56fcfcde09b7741196f121570cd8b2b969b9c02b859e13c8158865", "generation_seconds_batch": 18.48932155710645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3268353939056396, "target_z": -0.3320311766415681, "ce": 6.133810997009277, "genuineness": 4.68307638168335, "identity": 0.644281804561615, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ac4c4808c198fe50a302a413f18c87b81aff970490732d21f4c3bbd424f9a030.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "ac7563c7b939dcf5f634856913ae6245e6d890576828c3ed8813da0d0a2d6946", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e34c0f1f1ac319a01a3e82f4ab858bb2326660b780f73ebd0850de07be084cb9", "generation_seconds_batch": 22.70223337004427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.433278560638428, "target_z": 0.5351562218975161, "ce": 5.788304805755615, "genuineness": 4.796044826507568, "identity": 0.6520017981529236, "hypothesis": "Me he encontrado la cajita. Estaba junto a la. la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.11764705882352941, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ac7563c7b939dcf5f634856913ae6245e6d890576828c3ed8813da0d0a2d6946.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "ac80478cd3712c9e4884ff5609d389f9798298338c39b8941ca6e60d449d2c19", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "bb420c54118a960cd11971e182277f863b605908920d38b7f96ad202235b0b78", "generation_seconds_batch": 17.520284128026105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.848906993865967, "target_z": 0.07714836934235247, "ce": 5.569881439208984, "genuineness": 4.482243061065674, "identity": 0.5786147117614746, "hypothesis": "La puerta sigue abierta y he dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ac80478cd3712c9e4884ff5609d389f9798298338c39b8941ca6e60d449d2c19.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "acc9d1d42ca1f95468c188d46f0fd86929103714c55ac78bd220346a6534f25d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1715896f2c872726f70faf649d52f4eb2f1e341fc6965cf2d9911a66c0fc8d6e", "generation_seconds_batch": 22.848537919926457, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.433278560638428, "target_z": 0.5351562218975161, "ce": 5.752562999725342, "genuineness": 4.607763767242432, "identity": 0.7294180393218994, "hypothesis": "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_CASU_reg/acc9d1d42ca1f95468c188d46f0fd86929103714c55ac78bd220346a6534f25d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "accfd778670111035550ddad4cf9448f182d5602edc08e1caead35149780f8cc", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "5d7be8e594d6cf109a75f07ee612c2a9cd885715cecf2bb56ea9b9c3939bec35", "generation_seconds_batch": 17.76293143199291, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.10184383392334, "target_z": 0.27539049050954156, "ce": 5.974957466125488, "genuineness": 4.2374773025512695, "identity": 0.3761810064315796, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/accfd778670111035550ddad4cf9448f182d5602edc08e1caead35149780f8cc.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "ace4dda0bb378fd1c7a3d5666988cf99be72b242584f5e1bed207843c05f54cf", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "40217d3ad00590009e5539c3648306a2fec771c2109f8828adc33d65e640b12e", "generation_seconds_batch": 18.006599662010558, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.781934976577759, "target_z": 0.02465829014689602, "ce": 6.1894097328186035, "genuineness": 4.111956596374512, "identity": 0.2627217471599579, "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_CASU_reg/ace4dda0bb378fd1c7a3d5666988cf99be72b242584f5e1bed207843c05f54cf.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "ad07f3d9384e675b14ca7fade7a095bef78423be0e7a1584a52fedf8fd189b0c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "1af220311888fe4263ed0e3e521ce8a9763f5121ce9bf25dc192e6698299a05f", "generation_seconds_batch": 17.416352033033036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6539981365203857, "target_z": -0.8593750189783008, "ce": 6.070269584655762, "genuineness": 4.582659721374512, "identity": 0.6586111783981323, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ad07f3d9384e675b14ca7fade7a095bef78423be0e7a1584a52fedf8fd189b0c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "ad5ef2694725670104bed77d19255005b554185e06eced942d42af3caff6f75a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "833f480544b254c84857c39e4951e7b02f6aa7a4e9a5c60314e9237c8da3e5fd", "generation_seconds_batch": 25.922974191023968, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.54986834526062, "target_z": -0.157226567883268, "ce": 6.125868320465088, "genuineness": 4.65797233581543, "identity": 0.7636043429374695, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ad5ef2694725670104bed77d19255005b554185e06eced942d42af3caff6f75a.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "e135c482d3fe821d349a8b2faa6ee900a0ece1ddabc79b5f1927959220f3c75e", "generation_seconds_batch": 16.215560136013664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.242107629776001, "target_z": -0.3984375240878433, "ce": 6.157639026641846, "genuineness": 4.375550270080566, "identity": 0.5385793447494507, "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_CASU_reg/adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "f0a31a22dba7ac89b61cfba53f3cb8e5b798c366cb093cf5c5f730fe265b38f1", "generation_seconds_batch": 15.742170464014634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.51444673538208, "target_z": -0.9687500145986929, "ce": 6.340320587158203, "genuineness": 4.15588903427124, "identity": 0.3972666561603546, "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_CASU_reg/ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258.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": "aea1362d1a09aead5ffa4c3a7f76cedfd293384314a36f5fb229275c150dfce4", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "dc8e605234a69cf9e853ddaa46f562ef1a434f77d1aab1eb8180b22e3965cc03", "generation_seconds_batch": 22.545689053018577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.092588424682617, "target_z": -0.5156248992690189, "ce": 6.1894097328186035, "genuineness": 4.532451152801514, "identity": 0.6109720468521118, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/aea1362d1a09aead5ffa4c3a7f76cedfd293384314a36f5fb229275c150dfce4.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "af1ab8909966f7d00758fd651ca9c9e7e66b20b90079bdd7ceacd7ea778c3e48", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a03cea713e0b7287b1673122e360e0ba92c517134017c4b36e229dd82302b0eb", "generation_seconds_batch": 25.41339203598909, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.556098222732544, "target_z": -0.15234383065777832, "ce": 6.109982967376709, "genuineness": 4.607763767242432, "identity": 0.7158790826797485, "hypothesis": "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_CASU_reg/af1ab8909966f7d00758fd651ca9c9e7e66b20b90079bdd7ceacd7ea778c3e48.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "af49652cdc522e3367f56118c2c39f70d1dcf91f12b06d0b67bbba87f93dca8a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7b710e1c4adb0cea1f6e9904647df19d1c69713112a58aeae825469527ecf606", "generation_seconds_batch": 21.84295719896909, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.980982542037964, "target_z": 0.180664080292157, "ce": 5.458683967590332, "genuineness": 4.6203155517578125, "identity": 0.6163121461868286, "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_CASU_reg/af49652cdc522e3367f56118c2c39f70d1dcf91f12b06d0b67bbba87f93dca8a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "af67af0ae0fba6ee75fb9cfea575851503b2ccc023c2bec0c4830cd24d325499", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "31456e8fe7767b5396bc30d21b6476776b091bcfe8e186dcb519f056b15f7e9a", "generation_seconds_batch": 22.92002897092607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8644819259643555, "target_z": 0.08935539926934594, "ce": 5.998785495758057, "genuineness": 4.331617832183838, "identity": 0.47625237703323364, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/af67af0ae0fba6ee75fb9cfea575851503b2ccc023c2bec0c4830cd24d325499.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": "afbac1bdcca36ca76bba377c9dddb0c9b1b3b812b28978b948b8bc911806e6e6", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "20bffef71f57855f29401e155e03477e22560e1b8b0a40fde995aa997ae8d5b2", "generation_seconds_batch": 18.27532177802641, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7993788719177246, "target_z": 0.03833014124153638, "ce": 5.871703147888184, "genuineness": 4.557555198669434, "identity": 0.6819698214530945, "hypothesis": "Las llaves están en la mesa de la cocina, miren el cajón, y luego junto a la ventana, llévatelas cuando te vayas.", "wer": 0.08695652173913043, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/afbac1bdcca36ca76bba377c9dddb0c9b1b3b812b28978b948b8bc911806e6e6.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "b0369dd60a23da6b9ce2d3ab736cb9da4bbc750c4c6d3755810d41234a031774", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "37f5565dd508b0e8e2f92c7318b74a888887374556415151caea05d88ff17ede", "generation_seconds_batch": 24.344302712008357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8384053707122803, "target_z": -0.7148437547445752, "ce": 5.784333229064941, "genuineness": 4.557555198669434, "identity": 0.6294946670532227, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b0369dd60a23da6b9ce2d3ab736cb9da4bbc750c4c6d3755810d41234a031774.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "b04e05f3b58ec5fd0fab2d2eada66b9933d6bfc13eb0ac44c593c2fb02ecb1a6", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "a21ec7e835ba103895ce9ae6055adfced3d7825a3cc6afba66e74481a5c701e7", "generation_seconds_batch": 22.70243758999277, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0626845359802246, "target_z": -0.5390624116779079, "ce": 6.221180438995361, "genuineness": 4.570107460021973, "identity": 0.5042324066162109, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b04e05f3b58ec5fd0fab2d2eada66b9933d6bfc13eb0ac44c593c2fb02ecb1a6.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "b0b534e5313b3a5c618705aa1c5c5706116744fd70815faea8842dd87bc4bda5", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "18a940a6868b7e0d19e1491a0fc3faf1b8f09a546a894e0e8d89c500c893a0df", "generation_seconds_batch": 24.5633657600265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.166635513305664, "target_z": 0.32617170510771126, "ce": 5.8875885009765625, "genuineness": 4.494794845581055, "identity": 0.6240687370300293, "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_CASU_reg/b0b534e5313b3a5c618705aa1c5c5706116744fd70815faea8842dd87bc4bda5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "b1ff6c4115b0a85657352798e5cc6bb4f0563bffe399243e04bb1eaf79d17d83", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "ce05105b2be44639d046290cfd519a8ba97e615573bdb1018513d2df74526108", "generation_seconds_batch": 17.550878630019724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.788787841796875, "target_z": 0.03002930109493467, "ce": 6.0146708488464355, "genuineness": 4.444586753845215, "identity": 0.31835123896598816, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b1ff6c4115b0a85657352798e5cc6bb4f0563bffe399243e04bb1eaf79d17d83.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "b206719b52a81e2258ebf40fb65ec8ba79a1d565dd5883227ce20f5a36e660f3", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "743f728bc9f330257efa7fbeda0de1bc51dd66c7984fc993b9c624e4a8c3cc23", "generation_seconds_batch": 22.333664865931496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.251363277435303, "target_z": 0.3925780525539864, "ce": 5.6254801750183105, "genuineness": 4.68307638168335, "identity": 0.8615642786026001, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre eh la silla. Dime cuándo estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b206719b52a81e2258ebf40fb65ec8ba79a1d565dd5883227ce20f5a36e660f3.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "b2282783d3656bf3289265066f6c2e4a0b40ea70b584bc96d39c1383d24f1357", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "303302dc1d83de6121687c70ea0d066f20d07c29d53e1604daa8a7a90f07f8c2", "generation_seconds_batch": 15.978954939055257, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.087604284286499, "target_z": -0.5195312759126799, "ce": 6.038498878479004, "genuineness": 4.400654315948486, "identity": 0.690715491771698, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b2282783d3656bf3289265066f6c2e4a0b40ea70b584bc96d39c1383d24f1357.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "b239ffc705d99c41045ff15048ff638afb3a03f6b5c354cf0b88ce730d5629c3", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "147cdd1110dfb3b2d2174438b5090478e39f3b0ee95cd5488510ed274b2a519f", "generation_seconds_batch": 16.70463887299411, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6564009189605713, "target_z": -0.0737304532845097, "ce": 6.181467056274414, "genuineness": 4.645420074462891, "identity": 0.7501682639122009, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b239ffc705d99c41045ff15048ff638afb3a03f6b5c354cf0b88ce730d5629c3.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "b262b90134b300f889168cd07692939c8d4d6f30bbf55aecd2dd4576c239bbde", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "02cae9e64d4cd2640a8a2c70c431046764132ab6e6e7ca3ee6325361a4e1c9a7", "generation_seconds_batch": 22.779580950038508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8864654302597046, "target_z": -1.4609374948904574, "ce": 6.2767791748046875, "genuineness": 4.375550270080566, "identity": 0.6226004362106323, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b262b90134b300f889168cd07692939c8d4d6f30bbf55aecd2dd4576c239bbde.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": "b32d56bd598a94573ec183791ae761dbab1f40f175839651a3fcb4dfacad0b0b", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The keys are on the kitchen table.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "752233eef49b03ca553b50a294426c33a1e9f718e6789319f5f4926024dd77ba", "generation_seconds_batch": 20.834775227005593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.323631286621094, "target_z": 0.4492189255492822, "ce": 6.0305562019348145, "genuineness": 4.356721878051758, "identity": 0.4066397547721863, "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_CASU_reg/b32d56bd598a94573ec183791ae761dbab1f40f175839651a3fcb4dfacad0b0b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "b3c031e253b642fe87b32106b1958bbac4070cadc08921111dfbd7b2e75781a5", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "7850b669249f52a139d4b517d4ad969e46db5cac2c4e3aed328cb8f1838469ad", "generation_seconds_batch": 22.593999217962846, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.567845821380615, "target_z": 0.6406248408742473, "ce": 5.871703147888184, "genuineness": 4.632867813110352, "identity": 0.6658845543861389, "hypothesis": "La porte est encore ouverte J'ai posé ton manteau sur la chaise. Dis-moi quand quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b3c031e253b642fe87b32106b1958bbac4070cadc08921111dfbd7b2e75781a5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "b3c3dd8a95227f3c402fb679a2e19df396d044452984137727e6306afa2ba811", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "26e959fdef0633494c88425a4222e6ba92591892c99f455e60c268c422acd8b5", "generation_seconds_batch": 16.11165911704302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.597749710083008, "target_z": 0.6640623532831362, "ce": 5.95112943649292, "genuineness": 4.457139015197754, "identity": 0.3221041262149811, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b3c3dd8a95227f3c402fb679a2e19df396d044452984137727e6306afa2ba811.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": "b3e33169d692f307ec30a7e411928adcdc176ab1a2b4fff4eee0969d22ed90e2", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f49b55347b44964c0fd2813793d4c0c46e6ef8cacba96dea80471cfc925355a0", "generation_seconds_batch": 22.40629797708243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9480528831481934, "target_z": -0.628906271533072, "ce": 5.77639102935791, "genuineness": 4.457139015197754, "identity": 0.7545546889305115, "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_CASU_reg/b3e33169d692f307ec30a7e411928adcdc176ab1a2b4fff4eee0969d22ed90e2.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "f8eed474b269e4167d3cc974c5df47169315e70ccccdbdf4fab7dde022be2ea3", "generation_seconds_batch": 16.863939173985273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2769954204559326, "target_z": -0.3710938218985626, "ce": 6.308549880981445, "genuineness": 4.432034492492676, "identity": 0.48223501443862915, "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_CASU_reg/b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "b45acb3ddc8b735a99c7a38e3a2de65121b6b47e7569a17a35a3afc6ed412285", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "cc27b9c8f194d7f447930f08fff95dde4ea5fe2dff3280f0e9d6ce7c48dfe4a4", "generation_seconds_batch": 24.703866609022953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9772446155548096, "target_z": 0.17773443795686317, "ce": 6.213237762451172, "genuineness": 4.432034492492676, "identity": 0.2221137285232544, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b45acb3ddc8b735a99c7a38e3a2de65121b6b47e7569a17a35a3afc6ed412285.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "b47563b2def0ac753bf2cc7d84164ca72ea0d0b737fefcdf7ab38b511f719777", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "8ea8d474d7a01f009601dba9ede303d3c0714eab52e5189d078456e5ffa9c8b5", "generation_seconds_batch": 17.533436162979342, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.199031829833984, "target_z": 0.3515626861333346, "ce": 6.197352409362793, "genuineness": 4.319066047668457, "identity": 0.32613736391067505, "hypothesis": "I found the little box. It was beside the old window. We can look inside together?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b47563b2def0ac753bf2cc7d84164ca72ea0d0b737fefcdf7ab38b511f719777.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "b4993f3ea1c207e4a9b1fb044c7f5e9974a101fc5741a88aa9160edda0456f17", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "44f77b00c8bddd50eada2e6164f4a6ad66a6c729083aedcd9602b4acc4f9afbb", "generation_seconds_batch": 22.368857298977673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.831996917724609, "target_z": 0.8476562627738563, "ce": 6.006728172302246, "genuineness": 4.833701133728027, "identity": 0.7712564468383789, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b4993f3ea1c207e4a9b1fb044c7f5e9974a101fc5741a88aa9160edda0456f17.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "b49d67abd5752cf4d32541a4c274ab8688095aa47785f5e1c772fa2fe28d9fe6", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "63575c6907206eeeb113e0763fb9bb29bc41fd88c0bc011d00abfeaea46065bb", "generation_seconds_batch": 17.40407596004661, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5393667221069336, "target_z": -0.9492186919701957, "ce": 5.903473377227783, "genuineness": 4.582659721374512, "identity": 0.4389748275279999, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b49d67abd5752cf4d32541a4c274ab8688095aa47785f5e1c772fa2fe28d9fe6.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "b4c7132b9914a6cc7cd4fb1f32374bcfa49622cfe793118922ee17e4e8dd1d58", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c231cde5fd9d5be1fdb1766fcd686c53e6732e723be845d2d27a0886e0498523", "generation_seconds_batch": 21.760152954957448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.109320163726807, "target_z": 0.28125014890666766, "ce": 5.498397350311279, "genuineness": 4.369274139404297, "identity": -0.013428378850221634, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b4c7132b9914a6cc7cd4fb1f32374bcfa49622cfe793118922ee17e4e8dd1d58.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "b52edc33bc50242ebf2a8556edad02d6976672aecdcfd846e71acc6e33833ee6", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "9dd219f8b95b73e0bda1eeb5a63436f57775785ffe6e5fa545370f3886eb29a3", "generation_seconds_batch": 21.366885400027968, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.192267894744873, "target_z": -0.4374999824815685, "ce": 5.9431867599487305, "genuineness": 4.545003414154053, "identity": 0.6066722273826599, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b52edc33bc50242ebf2a8556edad02d6976672aecdcfd846e71acc6e33833ee6.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "b6a29ea3d5a7f83e8a76bccbef428eda583afbea6d4a2e577bea5369b14be5e9", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "1679f0d885cd197c291fc7d393ed4f4abfcb7dcdfdc5219215eedeb74cbf039f", "generation_seconds_batch": 25.694840676966123, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.987924814224243, "target_z": -0.5976561927001304, "ce": 6.1894097328186035, "genuineness": 4.68307638168335, "identity": 0.7709338665008545, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b6a29ea3d5a7f83e8a76bccbef428eda583afbea6d4a2e577bea5369b14be5e9.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": "b71a23f2f5d68611eac04104aebb6e897cdce8242e38313cc7abd3a5d7c24f05", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6524f1687c56e91654fddcf502fd51ada9de646df5cdac7e764795baa090466d", "generation_seconds_batch": 17.60428318497725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.936660289764404, "target_z": 0.9296873693416985, "ce": 5.990842819213867, "genuineness": 4.519899368286133, "identity": 0.5237109065055847, "hypothesis": "De weg gaat linksaf naar de brug. Loopt verder. Tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b71a23f2f5d68611eac04104aebb6e897cdce8242e38313cc7abd3a5d7c24f05.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "b7c674ed70f792946093076b7f7707d3b8296b04f2bf82a19a31143adad7781b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f5d61c36c390ff913a76848f6853944b89a3a5279d5aa62c69c89d7bac8ff470", "generation_seconds_batch": 22.382014365051873, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.433278560638428, "target_z": 0.5351562218975161, "ce": 5.6334228515625, "genuineness": 4.645420074462891, "identity": 0.6342301368713379, "hypothesis": "La porte est encore ouverte j'ai posé ton manteau sur la chaise Dis-mo dis-moi quand tu es prêt?", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b7c674ed70f792946093076b7f7707d3b8296b04f2bf82a19a31143adad7781b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "b867a1bd75a953ecf887a4c0b8bf2ac064b3b7b49b335151a79670ff99294b39", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "8e58d8ebaf5cb1ff83cd3640e93fae82849aa45fb55bace7cfecf718e9670639", "generation_seconds_batch": 24.344302712008357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.534916400909424, "target_z": -0.16894532408771248, "ce": 6.046441555023193, "genuineness": 4.582659721374512, "identity": 0.4759688377380371, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b867a1bd75a953ecf887a4c0b8bf2ac064b3b7b49b335151a79670ff99294b39.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "b87485c1a7577293fa3fb710bf4797a60ee80aef018284a04dc1af6322407a32", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "936ab083d21989dae4b94296921691c98d92b2b6c42d92c88206ed87299e4f2d", "generation_seconds_batch": 22.433138379012235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.577814102172852, "target_z": 0.6484375941615693, "ce": 5.7684478759765625, "genuineness": 4.293961524963379, "identity": 0.4580705165863037, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b87485c1a7577293fa3fb710bf4797a60ee80aef018284a04dc1af6322407a32.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": "b937775dc8e87da822ccb432bd2e4ae3603fca44ed323849e1ae9467d35db212", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Escribí el número en el sobre.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb4c625b11ab8a5fc83627ea11454e4fa1e1a2694c0447edcaf920b56a624aa4", "generation_seconds_batch": 15.098701035953127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.020854473114014, "target_z": 0.2119141591250987, "ce": 5.855817794799805, "genuineness": 4.6705241203308105, "identity": 0.8471297025680542, "hypothesis": "Se escribí el número en el sobre, léalo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b937775dc8e87da822ccb432bd2e4ae3603fca44ed323849e1ae9467d35db212.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "b98672badc2a52eb465ecf545074654209afcb8363ca06f536531c0b1e1ab1d2", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "7ba8f037bb7ef09d2c9837612126f228379d8b4587a984ff171ca6c341b210ae", "generation_seconds_batch": 23.245510738925077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.401595115661621, "target_z": -0.2734373956193457, "ce": 6.078212261199951, "genuineness": 4.65797233581543, "identity": 0.7581451535224915, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/b98672badc2a52eb465ecf545074654209afcb8363ca06f536531c0b1e1ab1d2.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "ba21328951f47481ff39117ec9ec06f91aee746f0dcf3309a0849a239af90b6f", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ee45b02a991be903d49f4ca5bf5b82046502e0a5ba4e05c9fa3a9996cfe415b", "generation_seconds_batch": 23.93859411496669, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0676684379577637, "target_z": -0.5351562218975161, "ce": 5.959072113037109, "genuineness": 4.469690799713135, "identity": 0.6792397499084473, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt ?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ba21328951f47481ff39117ec9ec06f91aee746f0dcf3309a0849a239af90b6f.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "a5770a60349f4cc3578f50a83fa7ecc99dc5b05b6fd41616e125ad4526a36890", "generation_seconds_batch": 16.508388701011427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1748239994049072, "target_z": -0.4511718335762089, "ce": 6.149696350097656, "genuineness": 4.582659721374512, "identity": 0.7414531111717224, "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_CASU_reg/ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "bb1309e58c46b756645dff2f339ccc5ec988df55a6dab59c9c00498e6d6b5435", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "771efe963361bcbc17e86c1e51a88f14c6d7f58baf38d2c24310c14157b90898", "generation_seconds_batch": 18.02974793803878, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.121779918670654, "target_z": 0.291015623357647, "ce": 5.927301406860352, "genuineness": 4.65797233581543, "identity": 0.7105497717857361, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/bb1309e58c46b756645dff2f339ccc5ec988df55a6dab59c9c00498e6d6b5435.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "bb411cdea86628870b3e9e5ea833946dce5b6abc6a98e08f9f346e96f7d91860", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c6bef85f19ee94e408463d3e32e10d409175da08d7883d674d69354b91cef471", "generation_seconds_batch": 22.877861893968657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2154078483581543, "target_z": -1.2031251386875828, "ce": 6.086154937744141, "genuineness": 4.394378662109375, "identity": 0.5053660869598389, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/bb411cdea86628870b3e9e5ea833946dce5b6abc6a98e08f9f346e96f7d91860.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "bb52ac04dcb91c78f1d484ba07a29fc517a28e7f1e917ff1e01ee9e83da21ce6", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "4b3c1ea662be7e6edf47dbee8e5f1e37db6cec146e5a7d442509e9bf503372e5", "generation_seconds_batch": 23.179981578025036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.268807411193848, "target_z": 0.40625009051189603, "ce": 6.24500846862793, "genuineness": 4.632867813110352, "identity": 0.2917240858078003, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/bb52ac04dcb91c78f1d484ba07a29fc517a28e7f1e917ff1e01ee9e83da21ce6.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": "bbb77e888e60cf937e43aca0114633651394f0193d12e74f76adaabae52bc32c", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The keys are on the kitchen table.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b7adb1071fab4705aaf5aaf63c370a68a0e348860fa12cc91d13cd5df5ec4c40", "generation_seconds_batch": 20.834775227005593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.73707914352417, "target_z": -0.010497978466437442, "ce": 6.221180438995361, "genuineness": 4.381826400756836, "identity": 0.48278796672821045, "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_CASU_reg/bbb77e888e60cf937e43aca0114633651394f0193d12e74f76adaabae52bc32c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "bbfe30fe5f272d24656d1ac0d218ca05177099394499f840a66966b9887e9331", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "0e9d658c96b399cb413e6df61bc675bb1cffaed64d77f18ee5f3d51d98f7d246", "generation_seconds_batch": 22.928676926065236, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.977956771850586, "target_z": -0.6054687591241831, "ce": 6.229123115539551, "genuineness": 4.469690799713135, "identity": 0.5819701552391052, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/bbfe30fe5f272d24656d1ac0d218ca05177099394499f840a66966b9887e9331.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "bc425ac90aaf6c88e39c2b2ad66d911a599853638b5d78b02e4a53f1bfb7fa96", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "ee16fb915594da2d8c291430517e829cfed5b580cb9f7f0752c1e61087287d0a", "generation_seconds_batch": 22.722531103063375, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.513022422790527, "target_z": 0.5976563795633996, "ce": 6.340320587158203, "genuineness": 4.582659721374512, "identity": 0.26176339387893677, "hypothesis": "I found the little box. It was beside the old window. We can look inside together?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/bc425ac90aaf6c88e39c2b2ad66d911a599853638b5d78b02e4a53f1bfb7fa96.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "bd370dc667ca310dd6aa74a91d003996d0326b2d0b61e4ead0680fcabc977b3e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "c72a726306c10f0b9b21bcd87bc5892b989569f5177f32ad4f0ad1cde4cbb777", "generation_seconds_batch": 22.557140112970956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6340622901916504, "target_z": -0.874999964963137, "ce": 5.903473377227783, "genuineness": 4.582659721374512, "identity": 0.7760336399078369, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/bd370dc667ca310dd6aa74a91d003996d0326b2d0b61e4ead0680fcabc977b3e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "bd815f660fa33c879adc4d7d63123ec460b2f0e088d09a30d81f0778fb843e2a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "b7f7cb4d628edcea1b1ac684e2b64e5147122613c1ca6c23847e53329cc450b1", "generation_seconds_batch": 22.701795726898126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8533573150634766, "target_z": -0.7031249985401307, "ce": 6.006728172302246, "genuineness": 4.394378662109375, "identity": 0.5701757669448853, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/bd815f660fa33c879adc4d7d63123ec460b2f0e088d09a30d81f0778fb843e2a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "bdb46dfc3db3855bf24754ca6d342b9f35015eba47056c48c464987655661131", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7483e88aa1234d13b90644a454b31c1c272a51d25a04900d9ab7f1c3119e243a", "generation_seconds_batch": 22.38563629903365, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.248871326446533, "target_z": 0.3906249576637906, "ce": 5.784333229064941, "genuineness": 4.293961524963379, "identity": 0.5997099280357361, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/bdb46dfc3db3855bf24754ca6d342b9f35015eba47056c48c464987655661131.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": "bde6ee8db5b1e4a0d4ac3951a5263d7f498ef8f7150aca81dc8e528c0221e073", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "24b59c21bc8de9ce649cf7dd477a28887dad26ff4055cfcbbaaa5a199a43b099", "generation_seconds_batch": 18.37050955905579, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.617685794830322, "target_z": 0.6796874861312417, "ce": 5.959072113037109, "genuineness": 4.557555198669434, "identity": 0.44135594367980957, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam, neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/bde6ee8db5b1e4a0d4ac3951a5263d7f498ef8f7150aca81dc8e528c0221e073.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "be24701a6ea10a4c1a32cf920fa56d25b3b919d3c1c0bc03c1fca02121eddbbc", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "0d61907ce33e23285cb827fea58ea24a06f84ef39a65b5b348ccf4ac32c17c46", "generation_seconds_batch": 23.110295646009035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.972972869873047, "target_z": -0.6093749489045748, "ce": 5.808161735534668, "genuineness": 4.6705241203308105, "identity": 0.7096396684646606, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/be24701a6ea10a4c1a32cf920fa56d25b3b919d3c1c0bc03c1fca02121eddbbc.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "be42b8e413a4e2eb38e64b9f7997a40eccf489fa1a0fb84d9661e1aca20d7cad", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "fd0a9d086ede871d88eab4f2dc56f1ef43b216236ee2bc4285b2b09b6529ea3a", "generation_seconds_batch": 16.664915927918628, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.358519077301025, "target_z": 0.47656262773856295, "ce": 5.506340026855469, "genuineness": 4.494794845581055, "identity": 0.7643827199935913, "hypothesis": "Me he encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/be42b8e413a4e2eb38e64b9f7997a40eccf489fa1a0fb84d9661e1aca20d7cad.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": "bea13dda8d59dc6b57aa7eb8595647fe9cbdc464b5667cf585cab30671b00623", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Someone moved the little blue chair.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1a2610016724ce9f3ece46a4f45575bdd194e4cfd3f29d9ede1ef06c7514f3ff", "generation_seconds_batch": 17.950054474989884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.919928789138794, "target_z": 0.1328125080292811, "ce": 6.268836498260498, "genuineness": 4.532451152801514, "identity": 0.3951537609100342, "hypothesis": "Someone m moved a little blue chair. It used to stand here beside the tall bookshelf. We can put it back after lunch.", "wer": 0.09090909090909091, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/bea13dda8d59dc6b57aa7eb8595647fe9cbdc464b5667cf585cab30671b00623.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "bee0a7d24f66af33502d6fdc210e48e79f60de9b2318be170a136321aeebbcb7", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "586530533bf6f6e90d20e4517d9dd10d1b9c68cb2207d42debf4f77ac9937203", "generation_seconds_batch": 23.679833487025462, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.228935241699219, "target_z": 0.37499982481568506, "ce": 5.593709468841553, "genuineness": 4.65797233581543, "identity": 0.5674779415130615, "hypothesis": "Me he encontrado la cajita, estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/bee0a7d24f66af33502d6fdc210e48e79f60de9b2318be170a136321aeebbcb7.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": "bf12601b38a18e585058bbd4d17509309bb7156f2fd2e82f404f0f81479d4a24", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "693799ac63dbb95661b66eb1e767ef6ca9a0620bf383e4b28126e7b8d2fc5858", "generation_seconds_batch": 25.32326394005213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.3186469078063965, "target_z": 0.445312362042352, "ce": 5.760505676269531, "genuineness": 4.570107460021973, "identity": 0.47474175691604614, "hypothesis": "De weg gaat linksaf naar de brug. Loopt verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/bf12601b38a18e585058bbd4d17509309bb7156f2fd2e82f404f0f81479d4a24.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "bf1386e9b9c56725a7a1758c5d24726035650d0918fc4b421bca186632138bf7", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "9f57ea82fe47e507aaa31a250918454efc4ff2ccc7b3ad304b7ec6fc426a6d1c", "generation_seconds_batch": 23.32972328993492, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.202235698699951, "target_z": -0.429687602920785, "ce": 6.046441555023193, "genuineness": 4.293961524963379, "identity": 0.2876976728439331, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/bf1386e9b9c56725a7a1758c5d24726035650d0918fc4b421bca186632138bf7.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "bf17f6b302ff76a0b3083dda461b30981b60de9a162a77c3fe4a5612e3c49afc", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "1dcf29125b43bba3302ed70418695429cbae96e9b795cee42e9ba304acf2715d", "generation_seconds_batch": 16.980974233942106, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.033314228057861, "target_z": 0.22167963357607806, "ce": 5.879645347595215, "genuineness": 4.7458367347717285, "identity": 0.8251721858978271, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise dis-moi, quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/bf17f6b302ff76a0b3083dda461b30981b60de9a162a77c3fe4a5612e3c49afc.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": "bf2a9fc37c61fac3da3fabdedfd2ce2f11f7ec20301ec9de710cc49807c09d54", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Your message arrived after I had left.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0aac3086aa6cf3e39d6a9dd394726ea978e6245dfe6e2adf3f431f7716b668fd", "generation_seconds_batch": 18.10422323900275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4987826347351074, "target_z": -0.19726557372209116, "ce": 6.316492557525635, "genuineness": 4.6705241203308105, "identity": 0.3518243432044983, "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_CASU_reg/bf2a9fc37c61fac3da3fabdedfd2ce2f11f7ec20301ec9de710cc49807c09d54.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "bf6f227417041b6e11bfd01a70e4d8e10bcef798734276921e8a9ae18b09e53b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "11e14b9cc451b986fb7f9c7e40a3b8f2ca2dfa23cccd500ea279d2407079d149", "generation_seconds_batch": 24.15367011295166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.681320905685425, "target_z": -0.05419913065601247, "ce": 6.09409761428833, "genuineness": 4.419482707977295, "identity": 0.7103208899497986, "hypothesis": "Ik heb het klein doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/bf6f227417041b6e11bfd01a70e4d8e10bcef798734276921e8a9ae18b09e53b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "c065c1a08718e0acd7e5bac5e7142519e30c472b6b7bd56866a9154ca38482b3", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "554594df7dce125ba02821a3c5d71aa9c569677fcd22631452616e51faf251cb", "generation_seconds_batch": 25.158052872982807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.975998640060425, "target_z": 0.17675789051176524, "ce": 6.078212261199951, "genuineness": 4.469690799713135, "identity": 0.5814763307571411, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c065c1a08718e0acd7e5bac5e7142519e30c472b6b7bd56866a9154ca38482b3.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "c0e493335167698f5d8356b28e785d9e43538e8d714177ec57af292344497ba8", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f4c9af03c0ac323a7053c885d34cb33bb19f6bf131900e5ef5feac1e54be1d79", "generation_seconds_batch": 22.661100882920437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6539089679718018, "target_z": -0.07568354817470557, "ce": 5.577824115753174, "genuineness": 4.632867813110352, "identity": 0.5326944589614868, "hypothesis": "Bueno la puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c0e493335167698f5d8356b28e785d9e43538e8d714177ec57af292344497ba8.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "c127b01d9aa2be87371124fdb3d65ea790733241c320357a559c704befa2ce77", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e425778ea05be38826a2643c95b6d2e4b180625c503df171d524ea9678da23ba", "generation_seconds_batch": 17.416352033033036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4845428466796875, "target_z": -0.9921875270075818, "ce": 6.213237762451172, "genuineness": 4.482243061065674, "identity": 0.5419713258743286, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c127b01d9aa2be87371124fdb3d65ea790733241c320357a559c704befa2ce77.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "c1c1fcacc582d7bb6b768f03d6a42532731e041ed48b4c5510c9d7aa6f7b4c47", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "0b88abd544f766feba16509173967e153b1aaa2c2609606d17ab6d969badfe78", "generation_seconds_batch": 22.687444000970572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8093466758728027, "target_z": 0.04614252080231987, "ce": 5.689021587371826, "genuineness": 4.457139015197754, "identity": 0.5077337026596069, "hypothesis": "La puerta sigue ya abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c1c1fcacc582d7bb6b768f03d6a42532731e041ed48b4c5510c9d7aa6f7b4c47.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "c1d2a73c009c062ed1ee7039367dc7c7d6d27b8b6ec3700952ca27418edd7875", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "85c4613eee475807f39abae03890329ecd24adb46d609697793ab6ef356615d1", "generation_seconds_batch": 18.400486216065474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.283759117126465, "target_z": 0.4179686598530713, "ce": 5.982900142669678, "genuineness": 4.356721878051758, "identity": 0.6229665279388428, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c1d2a73c009c062ed1ee7039367dc7c7d6d27b8b6ec3700952ca27418edd7875.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "c1d95980d528848c726c85b09ef7136b0b101a4f1e81be639b5a31907c146de4", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "35196404c3d456a8eb17a616a41e919938f496af0eb47773a0e2b7fb3b559a16", "generation_seconds_batch": 26.236191531992517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.712470769882202, "target_z": -0.02978507080202556, "ce": 6.403861999511719, "genuineness": 4.444586753845215, "identity": 0.40967079997062683, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c1d95980d528848c726c85b09ef7136b0b101a4f1e81be639b5a31907c146de4.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "c1e1acea7a42b9b2294efe51216cf855c4cef41d725b1d56ac85add2e4987bda", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "5d7ae8699f8dd7cd621c53f489991fc06afa1c734025b37318f1e65eb587decc", "generation_seconds_batch": 25.43309945904184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4247350692749023, "target_z": -1.0390625518253598, "ce": 6.4515180587768555, "genuineness": 4.532451152801514, "identity": 0.7007497549057007, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c1e1acea7a42b9b2294efe51216cf855c4cef41d725b1d56ac85add2e4987bda.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "c1e93411e174af6683d115f6574a7bb2b4a87c1cbf8ca42a7636a036beaa7424", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "89bdae2d78823ff49218b643981780321632f742212f036389097c00268f2db3", "generation_seconds_batch": 21.302939523011446, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.588494300842285, "target_z": -0.12695303649542422, "ce": 6.086154937744141, "genuineness": 4.482243061065674, "identity": 0.6789143085479736, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c1e93411e174af6683d115f6574a7bb2b4a87c1cbf8ca42a7636a036beaa7424.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": "c28bf0168fa30a83de02f5bd44215b6e3a836106e393d08f8654d506d2d843d9", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0aa3284b55b595cd3e74ca2c6cdab7e0b77d84dae8560ab2460543df3ca622f3", "generation_seconds_batch": 20.660011678934097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8594980239868164, "target_z": 0.08544920948895418, "ce": 5.681078910827637, "genuineness": 4.7081804275512695, "identity": 0.6486125588417053, "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_CASU_reg/c28bf0168fa30a83de02f5bd44215b6e3a836106e393d08f8654d506d2d843d9.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "c308e0c7ea443cf5953268da74aea3c16b6234c5e8daec996fcaf77dda6c0da9", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "8cf51984ee378ce993f2dfe480ea4b16ffe6869a24466821471c3245393a106f", "generation_seconds_batch": 21.795407379977405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2720115184783936, "target_z": -0.3750000116789543, "ce": 5.637393951416016, "genuineness": 4.482243061065674, "identity": 0.5303380489349365, "hypothesis": "Die Tür ist noch offen. Ich habe dein Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c308e0c7ea443cf5953268da74aea3c16b6234c5e8daec996fcaf77dda6c0da9.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "c3265acf16d245801ead86113038292889a5e707dd2b6aff605994025b972d13", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "3db1577d629f3eb56a84f5bc7526ea5e68e2b75118a97b60ca44233faa12fe3e", "generation_seconds_batch": 22.573456319049, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.274503469467163, "target_z": -0.37304691678875845, "ce": 5.831989288330078, "genuineness": 4.444586753845215, "identity": 0.3918710947036743, "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_CASU_reg/c3265acf16d245801ead86113038292889a5e707dd2b6aff605994025b972d13.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "c389d6bf27c130044cc0bac09f787263b591ef669a1b83e19ec7d5bf86eb8f2b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c894b52f2b72499c60d8ad5e055d43895c4fcae7dc5bc46396ab3b7e93333f81", "generation_seconds_batch": 25.850319700082764, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.58778190612793, "target_z": 0.6562499737223527, "ce": 5.283945083618164, "genuineness": 4.507347106933594, "identity": 0.6468809843063354, "hypothesis": "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_CASU_reg/c389d6bf27c130044cc0bac09f787263b591ef669a1b83e19ec7d5bf86eb8f2b.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "b7da9dcf415da26102e024d55780f1b8b3f9ec0f22542e5e942e6292123b4ccd", "generation_seconds_batch": 16.737210438936017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1748239994049072, "target_z": -0.4511718335762089, "ce": 6.125868320465088, "genuineness": 4.557555198669434, "identity": 0.5403609275817871, "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_CASU_reg/c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "c3d1eddb6d20f1fd1e8149fa65386af3dec520c065d1b5435b3b2ddbcf74cf9e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "fad7400bab7f1c2e5514f65cd3f29bcca7c110bcd131126936d4e1536e3e8dd2", "generation_seconds_batch": 22.853858230053447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.860743999481201, "target_z": 0.08642575693405212, "ce": 6.149696350097656, "genuineness": 4.6705241203308105, "identity": 0.6779248714447021, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c3d1eddb6d20f1fd1e8149fa65386af3dec520c065d1b5435b3b2ddbcf74cf9e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "c3fe44490fcf9b91de49d0779dea49353919b0ec1adf193e2f165b6d40794616", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "f3edf77aa3590459b7ceff41f66d10d12507831445eea070c398cb3073ea87d5", "generation_seconds_batch": 25.694840676966123, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.269519567489624, "target_z": -0.3769531065691502, "ce": 6.086154937744141, "genuineness": 4.607763767242432, "identity": 0.7636998295783997, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c3fe44490fcf9b91de49d0779dea49353919b0ec1adf193e2f165b6d40794616.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "c416b88e03686ed5ba54369a53f13eb9602a6c8f93636ea452d8299040a95e69", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a5ddb762597901071598389d832d63c3db5349d038bd76a29f10dccd1ccdd7a3", "generation_seconds_batch": 25.061136021045968, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7437098026275635, "target_z": -0.7890624817516338, "ce": 5.959072113037109, "genuineness": 4.582659721374512, "identity": 0.3794746398925781, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c416b88e03686ed5ba54369a53f13eb9602a6c8f93636ea452d8299040a95e69.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "c46b21f9f65cef82d32f1dd1957ea486e3d2ff0582a498124c2b463d27d78ac8", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "38c0ec2971ef1de64368ab2ce10c4b264be775c721fe04a608604358dfa4e949", "generation_seconds_batch": 23.047014658921398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2485160827636719, "target_z": -1.9609375416062749, "ce": 6.260893821716309, "genuineness": 4.193545341491699, "identity": 0.4557996690273285, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c46b21f9f65cef82d32f1dd1957ea486e3d2ff0582a498124c2b463d27d78ac8.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": "c48eb5ae868ea2de75348664a144c9e846685de975aac174f6efeed8fdb564f8", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I wrote the number on the envelope.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5f35736fc02162cc9860d758b6fd2f053e96f82a9a25cf87dfe17d8547f4f688", "generation_seconds_batch": 21.512020990019664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8115272521972656, "target_z": 0.04785157226287588, "ce": 6.141753673553467, "genuineness": 4.507347106933594, "identity": 0.4125051498413086, "hypothesis": "I r wrote the number on the envelope, read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.043478260869565216, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c48eb5ae868ea2de75348664a144c9e846685de975aac174f6efeed8fdb564f8.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "1388db4d1d9f96b2b2df3433dbe81f078a863fa628a359ae0883cb0fef3b9b0b", "generation_seconds_batch": 16.888734317035414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6345958709716797, "target_z": -0.09082040730026208, "ce": 6.348263263702393, "genuineness": 4.444586753845215, "identity": 0.5591088533401489, "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_CASU_reg/c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "c4cf24dc25e7c9a170ba213bad5ead09111dd6be19a6f38dfe36045fcfcd6a3d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a0c80ddd1bbe560f84255bf8c3c287a9c8831fa4ca025af44945d0413d56e073", "generation_seconds_batch": 25.101643697009422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7557690143585205, "target_z": 0.004150420073300854, "ce": 6.109982967376709, "genuineness": 4.7081804275512695, "identity": 0.5575563907623291, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c4cf24dc25e7c9a170ba213bad5ead09111dd6be19a6f38dfe36045fcfcd6a3d.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": "c4e39b91b020996eee94964381375d18352b76122a7e378e5bcd14531bce0425", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6e49f70b8ef1fd8e31dcd18dd4f9762da73875ab57557a0095240c1b1344a1fd", "generation_seconds_batch": 20.494089017040096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.772189140319824, "target_z": 0.8007812379560784, "ce": 5.903473377227783, "genuineness": 4.570107460021973, "identity": 0.702217698097229, "hypothesis": "Die bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg wanneer we kunnen praten.", "wer": 0.13043478260869565, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c4e39b91b020996eee94964381375d18352b76122a7e378e5bcd14531bce0425.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "c4e459645f31b0e86a8237e53ee54d61e1e8a69348954d582330ebb1ed977314", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "54c89b187dc444bee305992d81b14a313df95b410d5e3918274c3987c9e36acf", "generation_seconds_batch": 16.796318154083565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4489426612854004, "target_z": -0.2363282189790856, "ce": 6.09409761428833, "genuineness": 4.306513786315918, "identity": 0.22317279875278473, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c4e459645f31b0e86a8237e53ee54d61e1e8a69348954d582330ebb1ed977314.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "c50989f42b9e78df5d7a4482c1bb47d01b952d1a4725f852efe9e6f55c3a5bd5", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "57f314cc81bfa62f880f04acb377a7834edb9836e2a5a242b9bbd8c5f9431579", "generation_seconds_batch": 22.84490949800238, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.095792531967163, "target_z": -1.2968750014598693, "ce": 6.229123115539551, "genuineness": 4.381826400756836, "identity": 0.6171019077301025, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c50989f42b9e78df5d7a4482c1bb47d01b952d1a4725f852efe9e6f55c3a5bd5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "c51d262b468e304bdcb971069d61e2beed21873c0ad60ccde82a3c6b6bc66e6b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "73a69a6cde64b260c1d1977d16a4fb601563de57671f70dd4e8b2b16a9c5d837", "generation_seconds_batch": 17.887503511970863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0559206008911133, "target_z": -1.328125080292811, "ce": 6.372091293334961, "genuineness": 4.65797233581543, "identity": 0.7440536022186279, "hypothesis": "Ich habe die kleine Schacht gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c51d262b468e304bdcb971069d61e2beed21873c0ad60ccde82a3c6b6bc66e6b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "c5615cce98cd0d169827a60d42de39ac84ed103c1e8da80cb3bd9c53f539b294", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79c6af1889054fe56cbaadb2e86c5351d3704460146961428d207f52dd6213f6", "generation_seconds_batch": 24.17917200003285, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9161908626556396, "target_z": 0.1298828656939873, "ce": 5.863759994506836, "genuineness": 4.519899368286133, "identity": 0.4507787823677063, "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_CASU_reg/c5615cce98cd0d169827a60d42de39ac84ed103c1e8da80cb3bd9c53f539b294.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "c654b781e901a3e2d89f0a3056f05d1f4674d3185a51cd04080543c4c6d1beb7", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "04798e5701f6cea49c55e4a69e39a89591073a026cab376ad8fee3687e4b2d23", "generation_seconds_batch": 24.28665100398939, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.42331075668335, "target_z": 0.5273438423367327, "ce": 5.689021587371826, "genuineness": 4.821148872375488, "identity": 0.5311871767044067, "hypothesis": "Ni he encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c654b781e901a3e2d89f0a3056f05d1f4674d3185a51cd04080543c4c6d1beb7.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "c73581ea7178998b04cdcb2492b25fa945ca45242767e0e05c5697e76f824fea", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "c470a95ed179239f2489f185a409b9c56c92412a700d510a8ad57539622f8def", "generation_seconds_batch": 22.234044675016776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7586617469787598, "target_z": -0.7773437255471893, "ce": 5.919358730316162, "genuineness": 4.344170093536377, "identity": 0.6695519685745239, "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_CASU_reg/c73581ea7178998b04cdcb2492b25fa945ca45242767e0e05c5697e76f824fea.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "c75d3bfbebe3433b8f198a0da8e89e8be1957418a62c3d3a1426a6ffdd424a36", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "9ac2e10cf76aa0589123b87e4335edbb1a06e155f3ced65d3731556847b799e1", "generation_seconds_batch": 22.46121546404902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1947598457336426, "target_z": -0.4355468875913726, "ce": 6.252951145172119, "genuineness": 4.319066047668457, "identity": 0.41542065143585205, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c75d3bfbebe3433b8f198a0da8e89e8be1957418a62c3d3a1426a6ffdd424a36.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "c77c1dc51103f9eb71aba7dd8c338c3caaa2ff0314a9322d4a5b36d3a627eb4f", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "2f9258ef48f78ff6f0fa875e5b41e0c5161028bf64b59e5edfe6e7405011cd77", "generation_seconds_batch": 23.224070479045622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.448230743408203, "target_z": 0.5468751649652299, "ce": 6.395919322967529, "genuineness": 4.595211505889893, "identity": 0.3193534016609192, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c77c1dc51103f9eb71aba7dd8c338c3caaa2ff0314a9322d4a5b36d3a627eb4f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "c7faf168d0463952f40d8fb7dedeeb87cdbaab12a777eae74ecb09d4de320025", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "353d2463456549cb076534d27971dbec975fbc0dfd0b519b7c89dec34f4cc74f", "generation_seconds_batch": 22.788602139102295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.87631893157959, "target_z": 0.09863278686104558, "ce": 6.1020402908325195, "genuineness": 4.432034492492676, "identity": 0.7013255953788757, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c7faf168d0463952f40d8fb7dedeeb87cdbaab12a777eae74ecb09d4de320025.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "c814370be0cf8aec5ecca9a4571f5bfe344ad5f3ef10cb2e67ac858f66d4579a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "674592cd0c0655fb4a46e4808eb90b2c787f8e7873e14026c5b7dfd5ece8f0fe", "generation_seconds_batch": 22.007058573886752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9530367851257324, "target_z": -0.6250000817526803, "ce": 5.990842819213867, "genuineness": 4.758388519287109, "identity": 0.7124326825141907, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c814370be0cf8aec5ecca9a4571f5bfe344ad5f3ef10cb2e67ac858f66d4579a.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "bbcf94cee144fd2e53033715f8738a7c1b218ff388bcf3db32676cb0edc1e2b2", "generation_seconds_batch": 16.215560136013664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.268807411193848, "target_z": 0.40625009051189603, "ce": 6.24500846862793, "genuineness": 4.482243061065674, "identity": 0.6552877426147461, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Wanneer we kunnen praten.", "wer": 0.08695652173913043, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "c86a1e45eb72c026b49fc38435797d6bcb26491f64931d71e7100a957d40334b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "98ad53b3f40f818c743e67896dab6073f8ed35b4ff5dd41757b6b4e73f206d2e", "generation_seconds_batch": 21.827974291983992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.659515857696533, "target_z": -0.07128908467176485, "ce": 5.752562999725342, "genuineness": 4.231201171875, "identity": 0.11009549349546432, "hypothesis": "The door's still open. I put your coat on the chair. Tell me when you are ready.", "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_CASU_reg/c86a1e45eb72c026b49fc38435797d6bcb26491f64931d71e7100a957d40334b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "c8890d85c1e217dd218d86c83c69c8175178c5324cbbcd7395e9474397687673", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "83db968c8069bee70f9aeeddcbea97596b80268ef4136365e62812759b0d7b44", "generation_seconds_batch": 19.52921697194688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.519430637359619, "target_z": -0.9648438248183011, "ce": 6.0305562019348145, "genuineness": 4.419482707977295, "identity": 0.5312619805335999, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c8890d85c1e217dd218d86c83c69c8175178c5324cbbcd7395e9474397687673.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "5197dbcb47275dfa5721994ae69fcaed5f0d987f8a80d753e4b405cb4925e23b", "generation_seconds_batch": 17.287815985037014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4040870666503906, "target_z": -0.2714843007291498, "ce": 6.340320587158203, "genuineness": 4.3881025314331055, "identity": 0.3250082731246948, "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_CASU_reg/c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d.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": "c8bee8059dddc5d014dd899aa8505d94b5a9c1f727a46c3ab3c61e3dc886ba30", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f3c54543f6c5f71504ec38583779eed4e4dc20a7074fbe49e9ffc991ae982ae3", "generation_seconds_batch": 23.06294050696306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.283759117126465, "target_z": 0.4179686598530713, "ce": 6.300607204437256, "genuineness": 4.457139015197754, "identity": 0.823789656162262, "hypothesis": "Ik schreef het nummer op die envelop, lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c8bee8059dddc5d014dd899aa8505d94b5a9c1f727a46c3ab3c61e3dc886ba30.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": "c8c7031a1637ac9e7b9400c44b77bb079af682fb11f2bba9337ce5928472dcdb", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cc8a7319f37342c26c1ba8bb9d31148baf4419216cb47fd8f67ce30e2aa7ccd2", "generation_seconds_batch": 17.278336070943624, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.681943893432617, "target_z": -0.0537108569334635, "ce": 6.0146708488464355, "genuineness": 4.7834930419921875, "identity": 0.48176825046539307, "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_CASU_reg/c8c7031a1637ac9e7b9400c44b77bb079af682fb11f2bba9337ce5928472dcdb.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "c8d22d5036c555708dcffb80fa280bec329650106524e9ca013e513130ee6b88", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2ed238759d383916e4c572deb105ac6b0e54660200b868c46ef93c56c3f5e12c", "generation_seconds_batch": 23.627107370994054, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8513991832733154, "target_z": 0.0791016510958176, "ce": 5.792276382446289, "genuineness": 4.243753433227539, "identity": 0.44807469844818115, "hypothesis": "The door still open. I put your coat on the chair? Tell me when you are ready.", "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_CASU_reg/c8d22d5036c555708dcffb80fa280bec329650106524e9ca013e513130ee6b88.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "c907ed81193f289c6335761874b75576ad2110fa7e34418cfa2d538cc25cca0a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "55b89e890af9c1120e6f5731ab41a1b17c233d995c63867e982c7956fdfe0446", "generation_seconds_batch": 23.29694223497063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7049946784973145, "target_z": -0.03564454233588243, "ce": 5.744620323181152, "genuineness": 4.68307638168335, "identity": 0.6501824259757996, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c907ed81193f289c6335761874b75576ad2110fa7e34418cfa2d538cc25cca0a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "c963eaa151dee3755ad3ac4d9ca5590898a09fa0815c9ca949c0d059dbb7a0cb", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "baeb25057c3c805f12ae6b32450817ae2ddedcd45fef99bc7647076b81290e7a", "generation_seconds_batch": 22.242438716930337, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.194047451019287, "target_z": 0.3476561226264044, "ce": 5.649308204650879, "genuineness": 4.6203155517578125, "identity": 0.7789963483810425, "hypothesis": "Me he encontrado la cajita, estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c963eaa151dee3755ad3ac4d9ca5590898a09fa0815c9ca949c0d059dbb7a0cb.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "c9a029cc2308929c5ff967f5049df0e90973eb9d10936ecfd680a7f4398eb87b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff20f1ec3c961055e5f7013411831a3c766204ab3b0e80e88223829bea06ad4d", "generation_seconds_batch": 26.76233570894692, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2752156257629395, "target_z": -1.1562501138698047, "ce": 6.300607204437256, "genuineness": 4.6203155517578125, "identity": 0.6671382188796997, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c9a029cc2308929c5ff967f5049df0e90973eb9d10936ecfd680a7f4398eb87b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "c9db896ac833f2f179b8a9363397aef3bd9be9d0961dd43b46c670e8557ed580", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "da03ff177de1980ade8ae70c155f9e2fa70a88b5b4f4bc187504aad52c3c3889", "generation_seconds_batch": 22.949997978983447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5698041915893555, "target_z": -0.14160162189843178, "ce": 6.086154937744141, "genuineness": 4.293961524963379, "identity": 0.4869762659072876, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/c9db896ac833f2f179b8a9363397aef3bd9be9d0961dd43b46c670e8557ed580.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "ca4b5f805e63489dd71369b148ea69c039907f8e9d5aab066649cf82730f6c2c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "209143727661de115c4db1857bb3794f224d8726104e12881005ac84ec35c1d9", "generation_seconds_batch": 16.11165911704302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.592765808105469, "target_z": 0.6601561635027445, "ce": 5.780362129211426, "genuineness": 4.532451152801514, "identity": 0.5454144477844238, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ca4b5f805e63489dd71369b148ea69c039907f8e9d5aab066649cf82730f6c2c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "ca5346f3e6ee8130514b5119e5a53ab195d1a418515c303eb50db994b4408c7d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7316c37cb25a665ce5ebeb389a9513bb33a19972ecc61e7dbeb1aacb70abab57", "generation_seconds_batch": 21.284669721964747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5892064571380615, "target_z": -0.9101562335764705, "ce": 6.372091293334961, "genuineness": 4.406930446624756, "identity": 0.6951545476913452, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ca5346f3e6ee8130514b5119e5a53ab195d1a418515c303eb50db994b4408c7d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "cabd1333ab6acbc9c0bfbefff31fd9eb9c81670afeeac22d26bba0d1a429cfd5", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "683479aa7f7fa9a51732d8e639926455dfb1a57ebfc5fdb1fdb9ce766501566f", "generation_seconds_batch": 17.434772856067866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.401595115661621, "target_z": -0.2734373956193457, "ce": 5.879645347595215, "genuineness": 4.281409740447998, "identity": 0.5929573178291321, "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_CASU_reg/cabd1333ab6acbc9c0bfbefff31fd9eb9c81670afeeac22d26bba0d1a429cfd5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "cb1212b0feb8a24aa884997d3a94102bdec16936d41a292326fefc613e0ba3fa", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bd174883a1426c7523255adf6554f82692c973f4ccce1db5da96d3b7a5c431db", "generation_seconds_batch": 23.857949275989085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.194047451019287, "target_z": 0.3476561226264044, "ce": 6.046441555023193, "genuineness": 4.457139015197754, "identity": 0.5946600437164307, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/cb1212b0feb8a24aa884997d3a94102bdec16936d41a292326fefc613e0ba3fa.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "cb262cc9c36c000376d4ae836089b086d9696bf817cf4f37a7962acf440b2aeb", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "3204c367f0392b42ff759787345a95f8b000decdde0a50ee9d5a1911a5eb4004", "generation_seconds_batch": 17.38991058699321, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1641435623168945, "target_z": 0.3242186102175154, "ce": 5.454712867736816, "genuineness": 4.570107460021973, "identity": 0.8158726692199707, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja, podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/cb262cc9c36c000376d4ae836089b086d9696bf817cf4f37a7962acf440b2aeb.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "cb392db1fddd13d37ea44d43371016e69d057c4dc17915b010792230a4dcd9bc", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1afacaf15a54c47839294d8b407d81a22d07e2cb63dffaea00c4a55100818b60", "generation_seconds_batch": 25.724504982936196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1591596603393555, "target_z": 0.32031242043712366, "ce": 5.80021858215332, "genuineness": 4.7081804275512695, "identity": 0.6684882640838623, "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_CASU_reg/cb392db1fddd13d37ea44d43371016e69d057c4dc17915b010792230a4dcd9bc.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": "cb711cc4913d60ac9e545264f20c8955fbeaed907f16671707bfece48b48850e", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "54599ea5942325f30ff7def6e9922b182c1f02a30d1e4fcfff8fea1b85ec3c49", "generation_seconds_batch": 17.036728291073814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1075401306152344, "target_z": -0.5039063299278437, "ce": 6.0305562019348145, "genuineness": 4.356721878051758, "identity": 0.6867059469223022, "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_CASU_reg/cb711cc4913d60ac9e545264f20c8955fbeaed907f16671707bfece48b48850e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "cb89d659dd657276078deb3f879350f346f486b583d4ff98b8d296965ced3c11", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fff9266ee6b058d12fa7082074d15166289ac4e3d9ab03cb8f6fe5df194688d0", "generation_seconds_batch": 24.606192229082808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.053250312805176, "target_z": 0.23730476642418355, "ce": 5.5539960861206055, "genuineness": 4.482243061065674, "identity": 0.6275845766067505, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/cb89d659dd657276078deb3f879350f346f486b583d4ff98b8d296965ced3c11.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "cbb8e0183dcd22f539d30987245a896fc0a20f151894611f29a9dab00038db5c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "195bf75cb5dc1902c5669d4092fc54fb7310dc09b2f861b828ee41dd867b859e", "generation_seconds_batch": 24.15367011295166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7683846950531006, "target_z": 0.014038103102369397, "ce": 6.109982967376709, "genuineness": 4.375550270080566, "identity": 0.7330684661865234, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/cbb8e0183dcd22f539d30987245a896fc0a20f151894611f29a9dab00038db5c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "cbd69f53bf39dfb9857a7e26c6702a31f5bd9f185027d27c29675fa92001181e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d79bdef3074e6dc5bebf8d61a16d3fca8f05ab10bd6fec3088da5b94e2aa13cb", "generation_seconds_batch": 17.177611363935284, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.388422966003418, "target_z": 0.500000140147452, "ce": 6.149696350097656, "genuineness": 4.6956281661987305, "identity": 0.23705154657363892, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/cbd69f53bf39dfb9857a7e26c6702a31f5bd9f185027d27c29675fa92001181e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "ccc93df6f8623d25baffc8d5e3e51ca45f69b496e901de4b8e9b7725597123aa", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "b34a23d578721b11c85a335baa4474d7508470fad50f0acc1eca50c9e8f35c8b", "generation_seconds_batch": 16.05221094901208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5698041915893555, "target_z": -0.14160162189843178, "ce": 6.109982967376709, "genuineness": 4.337893962860107, "identity": 0.6955271363258362, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ccc93df6f8623d25baffc8d5e3e51ca45f69b496e901de4b8e9b7725597123aa.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "cd221e2c8a5da1131688fb922538fb72f750266873541a503c456aa6518b42c8", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "270482098d2f58a987e8e4517198509d2ea536524b1519d030e8942918833407", "generation_seconds_batch": 16.514187742955983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.967988967895508, "target_z": -0.6132811386849666, "ce": 5.998785495758057, "genuineness": 4.6705241203308105, "identity": 0.7864630222320557, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/cd221e2c8a5da1131688fb922538fb72f750266873541a503c456aa6518b42c8.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "cd56cad9d94dccce16e6b92155c6a603d8935ed42c61e9e209e655a91c5373d1", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "3996dc2382bacfaff62c5ce7b12cadff86bdd0cfe99ea5284c21d935df92cc15", "generation_seconds_batch": 22.663959964062087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.812060832977295, "target_z": 0.8320311299257508, "ce": 5.998785495758057, "genuineness": 4.507347106933594, "identity": 0.6390999555587769, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/cd56cad9d94dccce16e6b92155c6a603d8935ed42c61e9e209e655a91c5373d1.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "cdf7a3a65be4342648e603625876a95da6178bdbf575373b340c0c2679f5efee", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f376b93f9c25a7db43d16073c4b292c3f6f1e6624a8a3315fe9361cd5498a760", "generation_seconds_batch": 23.129906447022222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 5.021388053894043, "target_z": 0.9960937167879736, "ce": 6.0305562019348145, "genuineness": 4.494794845581055, "identity": 0.5833362340927124, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/cdf7a3a65be4342648e603625876a95da6178bdbf575373b340c0c2679f5efee.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "ce3ecb97301f05e5e6e165ec75add34b70df431fa6a4d48e1a62aaadb5bbbcc3", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f106c04c6c7529167c94a7d1a6408e5fe9f88970f29d8a90d328a0bfa02fb3df", "generation_seconds_batch": 16.646278284024447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.498070240020752, "target_z": 0.5859374364956859, "ce": 5.95112943649292, "genuineness": 4.362998008728027, "identity": 0.6791107654571533, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ce3ecb97301f05e5e6e165ec75add34b70df431fa6a4d48e1a62aaadb5bbbcc3.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "ce43cd11b90d5c31d758b736015b1046b6b00b6d043de8098ce0a215fd2d515b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "a2aa90eaa19e670479d628c549b8c99eee6367b4e9d77a9f535e2a79b29ad229", "generation_seconds_batch": 16.632900099968538, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.67891788482666, "target_z": -0.8398438832130728, "ce": 6.284721851348877, "genuineness": 4.6203155517578125, "identity": 0.5754713416099548, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ce43cd11b90d5c31d758b736015b1046b6b00b6d043de8098ce0a215fd2d515b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "ce8235855f37c2866135893c30c29a1ee33dce8fa3e39c33afa01bc2e6c6f98d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "b3ad76d08177c08e8fb69a4e87f36f4ef309b64b04bc8b169309b76c9a6f0cb3", "generation_seconds_batch": 21.78946642507799, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.321139335632324, "target_z": 0.44726583065908637, "ce": 6.038498878479004, "genuineness": 4.457139015197754, "identity": 0.6084599494934082, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ce8235855f37c2866135893c30c29a1ee33dce8fa3e39c33afa01bc2e6c6f98d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "ce896930f20d3a86fbe1f092afa3ef2bd9b55af28182ca81f5e526afe9164c80", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "d9f2089d5e78efa804b30911b508a162aceb436cca2a5e51fd3a70f1e540ba4b", "generation_seconds_batch": 22.573456319049, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.28696346282959, "target_z": -0.36328125547450985, "ce": 5.375286102294922, "genuineness": 4.507347106933594, "identity": 0.5787893533706665, "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_CASU_reg/ce896930f20d3a86fbe1f092afa3ef2bd9b55af28182ca81f5e526afe9164c80.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "cea103550748f7a948bcc8029720fc03762111cc6ca99f334993e77bce1e27d9", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c27b871d726989e1420dcc7f293ca1b2ec319b0aa868ae094af2158029dbfa9b", "generation_seconds_batch": 22.74679440294858, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0128445625305176, "target_z": -0.5781250569349023, "ce": 6.165581703186035, "genuineness": 4.519899368286133, "identity": 0.7260359525680542, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/cea103550748f7a948bcc8029720fc03762111cc6ca99f334993e77bce1e27d9.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "ceba11d7eabfe9dd9c28fcebcbada3e159e9ca66bfb7aa3aa163e2cf273ae9cd", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "6bfa7b240b1fe951b9a9d65f1c0ea8b414f7344e7f8b16ad4352de259182fee8", "generation_seconds_batch": 22.726629375945777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.079416275024414, "target_z": 0.2578126364977787, "ce": 6.24500846862793, "genuineness": 4.720732688903809, "identity": 0.7197099924087524, "hypothesis": "J'ai trouvé la petite boîte. Elle était prête, la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ceba11d7eabfe9dd9c28fcebcbada3e159e9ca66bfb7aa3aa163e2cf273ae9cd.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "ceea3f6a5ef50894e957270bc82165881598b07bcf7c51fe985384402df96ef7", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "245acc6265e3700cf4a911d878e1b8d724f01f0543c3046379633755f9d193c0", "generation_seconds_batch": 17.550878630019724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.577814102172852, "target_z": 0.6484375941615693, "ce": 5.879645347595215, "genuineness": 4.400654315948486, "identity": 0.637629508972168, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ceea3f6a5ef50894e957270bc82165881598b07bcf7c51fe985384402df96ef7.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "cf3457146f21965e83da170f6d2b3f7f5c8daaaabcf4893ecab453cb4dfd2968", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "50bff008041db5db35022f8b9613b169be9d7293c794734fae91d8606b603474", "generation_seconds_batch": 23.316566491033882, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2769954204559326, "target_z": -0.3710938218985626, "ce": 5.526196479797363, "genuineness": 4.400654315948486, "identity": 0.6672604084014893, "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_CASU_reg/cf3457146f21965e83da170f6d2b3f7f5c8daaaabcf4893ecab453cb4dfd2968.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "3a9b8a6f3841e29933649aa92166b1370d0a1d7d2f9b9e8d92f4aceb965eae67", "generation_seconds_batch": 19.275168676045723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.205440044403076, "target_z": -1.2109375182483662, "ce": 6.340320587158203, "genuineness": 4.557555198669434, "identity": 0.70778489112854, "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_CASU_reg/cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510.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": "cfe20c3f53aa5728fcc66d742c550a8ea15cf60ceac05f9c2b9a897550cb69f6", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e1dec9cb11bd93836cfc273e601a6972c4d55000fe12154f5eebf9c33daa94fd", "generation_seconds_batch": 16.67027512995992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.228935241699219, "target_z": 0.37499982481568506, "ce": 5.80021858215332, "genuineness": 4.582659721374512, "identity": 0.8425678014755249, "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_CASU_reg/cfe20c3f53aa5728fcc66d742c550a8ea15cf60ceac05f9c2b9a897550cb69f6.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": "d023dc894cbf1f2ed9becf6aaa32902d7c0c99c6a7345037cb0774c19dbaf5df", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f5d7ec4990121ba572da9aa305977606147e3d6d9941ada5257bf41dc60ed40", "generation_seconds_batch": 23.274564691004343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.967988967895508, "target_z": -0.6132811386849666, "ce": 5.998785495758057, "genuineness": 4.469690799713135, "identity": 0.49019190669059753, "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_CASU_reg/d023dc894cbf1f2ed9becf6aaa32902d7c0c99c6a7345037cb0774c19dbaf5df.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "d07f8b3c7e1ea8844c8c9cac8da76745b78fa7091b38dbe883df865c0d10dab4", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "c9578376d8da010e438f8bde0f04fc38f276f770c5875230696bff8251780219", "generation_seconds_batch": 21.90097497892566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.67891788482666, "target_z": -0.8398438832130728, "ce": 6.24500846862793, "genuineness": 4.507347106933594, "identity": 0.5865743160247803, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d07f8b3c7e1ea8844c8c9cac8da76745b78fa7091b38dbe883df865c0d10dab4.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "d0a39bf64887a651786a47bd3c041e90832a39ba9591261622952f11698057fd", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "244556c4acecb20096ebc879329202c34652c9eb80bf2458a45e4c76b765113c", "generation_seconds_batch": 23.918496854952537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6372663974761963, "target_z": -1.6562500671539875, "ce": 6.324435234069824, "genuineness": 4.281409740447998, "identity": 0.691799521446228, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d0a39bf64887a651786a47bd3c041e90832a39ba9591261622952f11698057fd.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "d0c2164a9f869f006268b8306deea5041c4325b44ada97f1c97910b742aa765c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "84d2e65fee46c7154f49100c99b332ed6009f013ff2154a3c21c37da8845eff7", "generation_seconds_batch": 22.892663629027084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2122037410736084, "target_z": -0.4218750364967323, "ce": 6.284721851348877, "genuineness": 4.645420074462891, "identity": 0.5916100740432739, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d0c2164a9f869f006268b8306deea5041c4325b44ada97f1c97910b742aa765c.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "01a44625e28e7dc2d7ade0ebcfc730faac64bda4475063806e33f89e935b23c5", "generation_seconds_batch": 16.83135937305633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3748953342437744, "target_z": -1.0781250102190851, "ce": 6.292664527893066, "genuineness": 4.017816066741943, "identity": 0.4185333847999573, "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_CASU_reg/d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "d15850031f16690ee246b27b344378e66a045f9002e25bd22e6f60afba900723", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "1380da446b042692646105e8e0dbb92a275a1d9aa18b22d7cdc021c50da3795a", "generation_seconds_batch": 20.199001275002956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.326123237609863, "target_z": 0.4511720204394781, "ce": 5.903473377227783, "genuineness": 4.758388519287109, "identity": 0.6202960014343262, "hypothesis": "J' j'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d15850031f16690ee246b27b344378e66a045f9002e25bd22e6f60afba900723.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "cfca952e256f60ce124ec98f20eeb8a25b20a3618c3ac3ef8d643eb493424f9f", "generation_seconds_batch": 17.087034616968594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9031970500946045, "target_z": -0.6640625401464055, "ce": 6.1894097328186035, "genuineness": 4.406930446624756, "identity": 0.8277086019515991, "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_CASU_reg/d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "d2a485ae46c08265a07d09551be96cf0ba523f61767f5cdcb56fa8bfc76ee610", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "4c64c30720268abddca6b5e0dd0c4bf9fd96c8fcdbf1441c7ce081232d4f5672", "generation_seconds_batch": 23.800322041963227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5137345790863037, "target_z": -0.18554681751764668, "ce": 6.197352409362793, "genuineness": 4.432034492492676, "identity": 0.5412646532058716, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d2a485ae46c08265a07d09551be96cf0ba523f61767f5cdcb56fa8bfc76ee610.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "d35f1148f583938d72c7e4845b024bb8bd22f4b949e52613ea60517d0bdbbcad", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e918157a7c885eaf568208fea7bd069f8d8bf81340bd14b9f1c626af8da53dca", "generation_seconds_batch": 22.99366586294491, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.238903522491455, "target_z": 0.38281257810300706, "ce": 5.998785495758057, "genuineness": 4.444586753845215, "identity": 0.649635910987854, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d35f1148f583938d72c7e4845b024bb8bd22f4b949e52613ea60517d0bdbbcad.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "d38ffb2c584ee354e94da880170d07ed6e955fd6431d35eda598decf43c4c2e6", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "564cb91f864abc42fe9f3e3b98310a5ebce2bb3b5a753dfffa1ac6832aea063d", "generation_seconds_batch": 21.58761911804322, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.238903522491455, "target_z": 0.38281257810300706, "ce": 5.982900142669678, "genuineness": 4.582659721374512, "identity": 0.1433655321598053, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d38ffb2c584ee354e94da880170d07ed6e955fd6431d35eda598decf43c4c2e6.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "866c8f10b474225e1089a29d23afd40e0f29b117d4437244a3c6450bfba15628", "generation_seconds_batch": 14.784412681008689, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2353439331054688, "target_z": -1.1875000058394771, "ce": 6.038498878479004, "genuineness": 4.413206577301025, "identity": 0.6324084997177124, "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_CASU_reg/d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "d42a01223610ead0db01cb5a87ed5f5c3d5ac840c05cecf2381867d24928b046", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "245ab581a2ce3b62a8fc737d674aae96bb77909522da6fabadfa8bb96b20af4d", "generation_seconds_batch": 24.49523817107547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9929087162017822, "target_z": -0.5937500029197386, "ce": 5.839932441711426, "genuineness": 4.444586753845215, "identity": 0.6053449511528015, "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_CASU_reg/d42a01223610ead0db01cb5a87ed5f5c3d5ac840c05cecf2381867d24928b046.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "d4431a2d8ef92750e71de48abeca01ff070683c103a47a3a4c7704fda8b0659f", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "6f507e0a1e7fa6a302b8a3321712ebeee9aca7b030c0ec47c4a940fe9b33de7d", "generation_seconds_batch": 21.429744058987126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7468912601470947, "target_z": -0.0028076206204738796, "ce": 5.617537498474121, "genuineness": 4.507347106933594, "identity": 0.742972731590271, "hypothesis": "J'ai trouvé la 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_CASU_reg/d4431a2d8ef92750e71de48abeca01ff070683c103a47a3a4c7704fda8b0659f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "d4c7787fc504e0ce5f5305badff2aa35411c72ab6c10636ab797b2e13154559b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "83de140f222563554d763d4fe544b20a109baaee25064b9824c4acb1a43f5281", "generation_seconds_batch": 16.796318154083565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3617231845855713, "target_z": -0.3046874744522874, "ce": 6.046441555023193, "genuineness": 4.231201171875, "identity": 0.37690579891204834, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d4c7787fc504e0ce5f5305badff2aa35411c72ab6c10636ab797b2e13154559b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "d4f3c95fd10c1ab93e86e309a4e54856c7b18d6da91bf361b678b115007385c3", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "0fdb2c490fb949258233ebf522e26460b3d78e0b9e525ca08919b66ecb808dea", "generation_seconds_batch": 23.90744831704069, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.393406867980957, "target_z": 0.5039063299278437, "ce": 6.372091293334961, "genuineness": 4.519899368286133, "identity": 0.3144538998603821, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d4f3c95fd10c1ab93e86e309a4e54856c7b18d6da91bf361b678b115007385c3.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "d4f639f66e3917e5a3c0b33da720ec8f232ecfd3bfbb8dd6d24293777f610678", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "cc35dc5a5442f6a6e6a82fd0b3f7eb3279cc6b87693493108538ed0098119203", "generation_seconds_batch": 22.229210576973855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.351042747497559, "target_z": 0.47070296934143685, "ce": 6.022613525390625, "genuineness": 4.532451152801514, "identity": 0.645767867565155, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d4f639f66e3917e5a3c0b33da720ec8f232ecfd3bfbb8dd6d24293777f610678.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "d519e2866c9f47b2e15ca6023d690cd2dbeb51eb36b85b018eb7c939e26a66d8", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "24a5376fa073078344bc73aa4fb7e5b3aacd4017be55d28ff1666b762e259898", "generation_seconds_batch": 17.461880658054724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2994234561920166, "target_z": -0.35351559416026124, "ce": 6.109982967376709, "genuineness": 4.444586753845215, "identity": 0.5295506715774536, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d519e2866c9f47b2e15ca6023d690cd2dbeb51eb36b85b018eb7c939e26a66d8.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "d5d31e3f935a0e1c6b90de3f46dce376715d86466c1e0d7eb787fae158514a79", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "1e5596b15535a929cab64d9db96cb7a862900b64ad028a03799886ab5f61ef4c", "generation_seconds_batch": 24.32315170194488, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7867858409881592, "target_z": -1.5390625051095426, "ce": 6.419747352600098, "genuineness": 4.419482707977295, "identity": 0.5362056493759155, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d5d31e3f935a0e1c6b90de3f46dce376715d86466c1e0d7eb787fae158514a79.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "d610938f138dab9a86595a3e4fbad2f14dba8e79653018a9cc83388dee42b6ed", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "2f674487cf7fdb6b158aee252e7505fc0a446f80481d76efb71f8343a435c904", "generation_seconds_batch": 22.928676926065236, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.740194082260132, "target_z": -0.0080566098536926, "ce": 6.09409761428833, "genuineness": 4.444586753845215, "identity": 0.6081822514533997, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d610938f138dab9a86595a3e4fbad2f14dba8e79653018a9cc83388dee42b6ed.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "d6758cc1ca18a462d9ee5784d3cf635bc787ed7af462e079030e11dd076bdc51", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "b0bcea3404ac8ec2edc407b4a605835de49e482a0d8c5cb6314271d1171ccda0", "generation_seconds_batch": 22.227656582952477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.396610975265503, "target_z": -0.27734377226300666, "ce": 5.9431867599487305, "genuineness": 4.796044826507568, "identity": 0.6851003170013428, "hypothesis": "La porte est encore ouverte j'ai posé ton manteau sur la chaise dis-moi quand y est prêt.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d6758cc1ca18a462d9ee5784d3cf635bc787ed7af462e079030e11dd076bdc51.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "d707d0694cf30f5c1432d518a5ad2ef0550b9d149ce0daf53868d9c94c61a35a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c679305469c1c027a529e11c8a8267299768faf15bd9b0227895212e329fcd4f", "generation_seconds_batch": 23.66319642099552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7387256622314453, "target_z": -0.7929688583952949, "ce": 6.006728172302246, "genuineness": 4.570107460021973, "identity": 0.5754362940788269, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d707d0694cf30f5c1432d518a5ad2ef0550b9d149ce0daf53868d9c94c61a35a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "d75386e7ce32b862535be116cccf0386c762c0c89c7b69abf9dffe7c470bb984", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "72c37f26d557cf9f7f90a1939989fb174fe9ed1571694c5572f7f61f5bc43276", "generation_seconds_batch": 22.46121546404902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6440300941467285, "target_z": -0.8671875854023535, "ce": 6.125868320465088, "genuineness": 4.193545341491699, "identity": 0.4286344647407532, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d75386e7ce32b862535be116cccf0386c762c0c89c7b69abf9dffe7c470bb984.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "d771650177256bb98ccc20a7328c0516a94df5e139e3cc0168bebcd02b7960a9", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "9b088e5f83b128769b2008daa9a9e5ca4457a76db025b93d1950d261f4fa73c7", "generation_seconds_batch": 18.400486216065474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.807076930999756, "target_z": 0.8281249401453591, "ce": 5.542081832885742, "genuineness": 4.331617832183838, "identity": 0.14018520712852478, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d771650177256bb98ccc20a7328c0516a94df5e139e3cc0168bebcd02b7960a9.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "d7d0685b99f31717324c25e14dbba2ad3f116ec2bae64b4bf1c35172c1b804b2", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "14a25df072e97e00f7ed34590145a7e1392d74ca664eeef5ca39504b5d1ae6f1", "generation_seconds_batch": 22.91413032298442, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6775827407836914, "target_z": -0.05712895985457553, "ce": 6.141753673553467, "genuineness": 4.6203155517578125, "identity": 0.750633180141449, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d7d0685b99f31717324c25e14dbba2ad3f116ec2bae64b4bf1c35172c1b804b2.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "d82735412bc6d256f56deeefc5b3b877a198fcdd428d5f189185bf63a7cd250a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f8f1465ba48bb71bb908a4c2ecad2bea32245841aa6e0736c3812a77fef987fa", "generation_seconds_batch": 21.620567710953765, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.132460117340088, "target_z": -0.48437500729934646, "ce": 6.125868320465088, "genuineness": 4.65797233581543, "identity": 0.640420138835907, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d82735412bc6d256f56deeefc5b3b877a198fcdd428d5f189185bf63a7cd250a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "d8525a74a3b4ff8c661ebc2e94225274776671a2e0ec8ef442abede0c18d299b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cd71814ef773a1deba6b95772b675760e7f50dd0a8e2e0ff3e1fec76fada4ac4", "generation_seconds_batch": 24.354885635897517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7224385738372803, "target_z": -0.02197269124124207, "ce": 5.593709468841553, "genuineness": 4.532451152801514, "identity": 0.6443425416946411, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d8525a74a3b4ff8c661ebc2e94225274776671a2e0ec8ef442abede0c18d299b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "d8deeccc82de33a427773573034b604fbf952a9e3d0a0da8b8ce4edd458688dc", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "9b52fc32f87f9c9ff9c788418b34af83d5a06b250da70b07979fdbeeb904bcd0", "generation_seconds_batch": 22.208209252916276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.237123727798462, "target_z": -0.40234371386823503, "ce": 6.078212261199951, "genuineness": 4.532451152801514, "identity": 0.8021177053451538, "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_CASU_reg/d8deeccc82de33a427773573034b604fbf952a9e3d0a0da8b8ce4edd458688dc.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "d9438a38e7237fbb5e3f826437073b13ffde987df41a0895b2b217fb650ff613", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2583c632b78434763e5bc427b88b26bd6b4c49fa85e02d28074852ed78f1e1ab", "generation_seconds_batch": 16.74104550492484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.189063549041748, "target_z": 0.34374993284601263, "ce": 6.086154937744141, "genuineness": 4.758388519287109, "identity": 0.6538357734680176, "hypothesis": "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_CASU_reg/d9438a38e7237fbb5e3f826437073b13ffde987df41a0895b2b217fb650ff613.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "d98dd0e9bfd42945a4729d9cb8a7de744696c9d922a1e545da56987355b20d2c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "794b7a78476fa765c875292cee8d9af6fdeedc692d8a277d1e90bbe0d541343a", "generation_seconds_batch": 23.755018191994168, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6208901405334473, "target_z": -0.10156242919633938, "ce": 6.046441555023193, "genuineness": 4.344170093536377, "identity": 0.6521410942077637, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/d98dd0e9bfd42945a4729d9cb8a7de744696c9d922a1e545da56987355b20d2c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "da834d0deab100dc455fc21ec4c3ab0b668b2d68554a0fc9be9fe5c70b786c84", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "92ebe4d832333236ba460e024b50308ec35b52ea4662148eb66f2ae7172d99ea", "generation_seconds_batch": 20.998760886024684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.902484893798828, "target_z": 0.11914065693464074, "ce": 6.221180438995361, "genuineness": 4.331617832183838, "identity": 0.34468942880630493, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/da834d0deab100dc455fc21ec4c3ab0b668b2d68554a0fc9be9fe5c70b786c84.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "da99e1fd5738ecacc9a50e32df7637fe401e56bbe9334ff1531e6bfa0c3677bc", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c07a880449f2c9883ef7a03ed09b0b74253a078218f691cdc76510a197363073", "generation_seconds_batch": 23.410271295928396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4247350692749023, "target_z": -1.0390625518253598, "ce": 5.990842819213867, "genuineness": 4.005263805389404, "identity": 0.24348735809326172, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/da99e1fd5738ecacc9a50e32df7637fe401e56bbe9334ff1531e6bfa0c3677bc.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "8cb5fb8a78d317fe73c645dc86533c7cff085c6eefe00a5965385e7df3e80829", "generation_seconds_batch": 17.376892061904073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.504478931427002, "target_z": -0.9765623941594764, "ce": 6.4515180587768555, "genuineness": 4.268857479095459, "identity": 0.6628957390785217, "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_CASU_reg/db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "db284165ada66bb32bff2d3527b0783bf4d999bb29b5021eab123cba43ad873b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "ffa4fac4736e2493b3ff028583cb3dad60ffb6dc74719f3fc4c510231ac1973a", "generation_seconds_batch": 18.809754970949143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.648301839828491, "target_z": -0.08007819854091554, "ce": 6.133810997009277, "genuineness": 4.645420074462891, "identity": 0.651412308216095, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/db284165ada66bb32bff2d3527b0783bf4d999bb29b5021eab123cba43ad873b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "db47668544fd93dff3892de6d068c1bd6259139b8fb70fc09dffe6f505c07334", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "1a644e94b89e6d5e908badcd9c221cad29e44f1871ac04f8e1cc6dfff552d4d1", "generation_seconds_batch": 18.835496819927357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1673479080200195, "target_z": -0.45703130511006573, "ce": 6.046441555023193, "genuineness": 4.3504462242126465, "identity": 0.3117640018463135, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/db47668544fd93dff3892de6d068c1bd6259139b8fb70fc09dffe6f505c07334.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "dc7a9701f84333eee9e0b044c25500cc4caed7e47a935b01b7795b59a9de0848", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "00f13fb8405246caac72e4287933af125102f35970fafdbc899c9ac58e873758", "generation_seconds_batch": 21.366885400027968, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1349520683288574, "target_z": -0.48242191240915056, "ce": 5.903473377227783, "genuineness": 4.582659721374512, "identity": 0.5887537598609924, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/dc7a9701f84333eee9e0b044c25500cc4caed7e47a935b01b7795b59a9de0848.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": "dd280b3e0bb69241f3c0c63458587d6d7870d67018b29106b8e65ce684fcc472", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3bbf3d436595e8611d52c1a04d1c5a2555a9e64ce61e458330680d6d6b616c86", "generation_seconds_batch": 23.06294050696306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.853891134262085, "target_z": 0.08105474598601348, "ce": 5.967014789581299, "genuineness": 4.582659721374512, "identity": 0.4438578188419342, "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_CASU_reg/dd280b3e0bb69241f3c0c63458587d6d7870d67018b29106b8e65ce684fcc472.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "ddd7a4be4032e190cb26c9c7c6c3db538f3616af9951957e842e77c71a9f677d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "ecf9b17a70e8e240d5d17480c91ffbdeb5829747d1bb3899d5004be3c5d153b2", "generation_seconds_batch": 24.960537275997922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7088217735290527, "target_z": -0.8164063708041839, "ce": 6.300607204437256, "genuineness": 4.607763767242432, "identity": 0.6564723253250122, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ddd7a4be4032e190cb26c9c7c6c3db538f3616af9951957e842e77c71a9f677d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "ddfbbd7dd786f90e5f523942c431367113a8b88109dd3ed15610609667b80cd7", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f65230983e9d621d200e541c640206cf697990431a0ca05ed3fe681c7ceef9a5", "generation_seconds_batch": 23.535603750962764, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.197251796722412, "target_z": -0.43359379270117676, "ce": 6.3641486167907715, "genuineness": 4.607763767242432, "identity": 0.5739800333976746, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ddfbbd7dd786f90e5f523942c431367113a8b88109dd3ed15610609667b80cd7.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "de0bcbde640d78cbe7a01a35266fbcbfd362e98fcd28139d17773021769f3fee", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "8a2c5eab832cabcf8fd6a40f5f628fda75200b331a60b0c60d2a98baa099e0a0", "generation_seconds_batch": 22.681499610072933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.866884708404541, "target_z": 0.874999964963137, "ce": 6.117925643920898, "genuineness": 4.557555198669434, "identity": 0.719498872756958, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/de0bcbde640d78cbe7a01a35266fbcbfd362e98fcd28139d17773021769f3fee.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "deddb61eded614345c76d02424bcd740098a6be66e1355d493bd11b5d31a203e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b471c2798a35abf03ae710107233d94f7fab03de33de5ab59ae3095aae9e953f", "generation_seconds_batch": 22.299424338969402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9772446155548096, "target_z": 0.17773443795686317, "ce": 5.262103080749512, "genuineness": 4.796044826507568, "identity": 0.6271182298660278, "hypothesis": "encontrado la cajita. estaba junto a la ventana vieja. ¿podemos mirar juntos lo que hay dentro?", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/deddb61eded614345c76d02424bcd740098a6be66e1355d493bd11b5d31a203e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "df649ed03b255568df518a9e3f8646bffeb70e06662cf29ab2a7aa0ca868eefe", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "29266437317ba5986b9d140c5fd87d71545001e02c4cadc04d804c9ab6f64e73", "generation_seconds_batch": 22.47733011993114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.518006324768066, "target_z": 0.6015625693437914, "ce": 5.871703147888184, "genuineness": 4.375550270080566, "identity": 0.5957721471786499, "hypothesis": "Me he encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/df649ed03b255568df518a9e3f8646bffeb70e06662cf29ab2a7aa0ca868eefe.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "e02d5a7349a3a3a6e89ba17f4b977b43a583511a1402859a6143be0d82ec9393", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "cf30dcca0f7473d23aaa4841d7324a42aaf58fab7d7f5c758d4f4b00b7adfe73", "generation_seconds_batch": 21.631574525032192, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.321139335632324, "target_z": 0.44726583065908637, "ce": 6.284721851348877, "genuineness": 4.507347106933594, "identity": 0.41674065589904785, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/e02d5a7349a3a3a6e89ba17f4b977b43a583511a1402859a6143be0d82ec9393.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "e0ad192a1243b3e44c43025f8f1820a15d1adaed23f74d6ddde72660d0c6ab26", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "9c5ea86673eff3c9fa66d3b55ede60641a8e31db07766920cb876220e062cf23", "generation_seconds_batch": 23.245510738925077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5747883319854736, "target_z": -0.13769524525477078, "ce": 6.109982967376709, "genuineness": 4.720732688903809, "identity": 0.7420246601104736, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/e0ad192a1243b3e44c43025f8f1820a15d1adaed23f74d6ddde72660d0c6ab26.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "e0c79f7b33caf957588d8011026076598bc29c8763c6772a9683263cb0d186b8", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "8d60c98a8e5c1e12372a0bf52d286fc4437c8ddc4771b8aa1af19cc0a0176f2e", "generation_seconds_batch": 17.887503511970863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9363051652908325, "target_z": -1.4218750364967323, "ce": 6.3641486167907715, "genuineness": 4.406930446624756, "identity": 0.5773729085922241, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/e0c79f7b33caf957588d8011026076598bc29c8763c6772a9683263cb0d186b8.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "e0d1b151355894b5f3d1cb9f5ad720564f347f0ffd5a2dc5ca0ea288e4cd37b9", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "b03f5ed89955e6af1526402a6071de28d1ed271c7f64d50e1df2380c081cefb8", "generation_seconds_batch": 17.290432508918457, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.017116546630859, "target_z": 0.2089845167898049, "ce": 5.895530700683594, "genuineness": 4.595211505889893, "identity": 0.5537370443344116, "hypothesis": "Die Tür 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_CASU_reg/e0d1b151355894b5f3d1cb9f5ad720564f347f0ffd5a2dc5ca0ea288e4cd37b9.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": "e1ec991a6f0985a923824eaab3d5bcf28940ee978073d268c49316747040678b", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fc5154f343903efc3b8337ea79ef83b851788bfde6aafa06866983322ab8cde4", "generation_seconds_batch": 17.345418073935434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6345958709716797, "target_z": -0.09082040730026208, "ce": 5.689021587371826, "genuineness": 4.720732688903809, "identity": 0.8586555123329163, "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_CASU_reg/e1ec991a6f0985a923824eaab3d5bcf28940ee978073d268c49316747040678b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "e2b0de1ff2a4ba537c5489db8d54d5bfefda2c73225ea9c1e60721a2279ce553", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0b2c5db31b9c584817c3ecde0e0dd35776e118a96c2969474bcc5c5f36206a10", "generation_seconds_batch": 23.012881717993878, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.990950584411621, "target_z": 0.18847664671620973, "ce": 5.804190158843994, "genuineness": 4.570107460021973, "identity": 0.6260371208190918, "hypothesis": "La porte est encore ouverte , j'ai posé ton manteau sur la chaise , dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/e2b0de1ff2a4ba537c5489db8d54d5bfefda2c73225ea9c1e60721a2279ce553.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "e2c6d893d23b46829adf02774c69f0e23c11ca38dab22d305fa1505a27aab8be", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "5841860ceffbccc54936ccf75b05e86aa9bd0f121546cad42e67d8f2ea2ba932", "generation_seconds_batch": 16.840930305072106, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.673933982849121, "target_z": -0.8437500729934646, "ce": 6.181467056274414, "genuineness": 4.055472373962402, "identity": 0.46248647570610046, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/e2c6d893d23b46829adf02774c69f0e23c11ca38dab22d305fa1505a27aab8be.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "e30c09150bb7cfc8cc28f798975786c9e655f395ede4afd891566f8da9388a0f", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "ba4e7a3837867588d10a2404610076edb6876938a91f4b87b2a1cf3f2e55ea20", "generation_seconds_batch": 21.825582813005894, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5212104320526123, "target_z": -0.17968753284705904, "ce": 5.839932441711426, "genuineness": 4.381826400756836, "identity": 0.6341533660888672, "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_CASU_reg/e30c09150bb7cfc8cc28f798975786c9e655f395ede4afd891566f8da9388a0f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "e33a153535468ddb09e74618d4e37b4a03be2f6f5f2c90a3992f9ec82b67e3a4", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "1ab6833909da148d71f46b6caa855aa8643349b4662c2931b8e21906eab752fb", "generation_seconds_batch": 24.977417498943396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.880056858062744, "target_z": 0.10156242919633938, "ce": 6.070269584655762, "genuineness": 4.6203155517578125, "identity": 0.6993083953857422, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/e33a153535468ddb09e74618d4e37b4a03be2f6f5f2c90a3992f9ec82b67e3a4.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "e3d6f0e8e4122a2bc06f45c63ac91d07008daeb39c9f89aa9b162c0094e186db", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "ad2b7027adf6360fc2f704db8ac032138737bec578f7011da60623c36350f698", "generation_seconds_batch": 21.70309393608477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.5279741287231445, "target_z": 0.6093749489045748, "ce": 5.982900142669678, "genuineness": 4.356721878051758, "identity": 0.5066425800323486, "hypothesis": "I f found the 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_CASU_reg/e3d6f0e8e4122a2bc06f45c63ac91d07008daeb39c9f89aa9b162c0094e186db.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": "e3ea90a3c232cdc5d9916a71b7c59e4479b92bc2d1b07da1cf38fff88b845b20", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "585baeb66f1a1762b1cf101e278c087ac92d29e781b80f7107edcbfa1b734cf0", "generation_seconds_batch": 17.089458236005157, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.032780647277832, "target_z": -0.5624999240867968, "ce": 5.649308204650879, "genuineness": 4.645420074462891, "identity": 0.8140827417373657, "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_CASU_reg/e3ea90a3c232cdc5d9916a71b7c59e4479b92bc2d1b07da1cf38fff88b845b20.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "e3ebf0cf23da140cc3ca6986f0c51012966f6b02bca1ac151974bbaa479dd30a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2cfe44d2af21e177809cbaaa0783e49244133efbc6aa041cd95e49965f59af0a", "generation_seconds_batch": 21.246189445024356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0060808658599854, "target_z": -1.3671875386865362, "ce": 6.300607204437256, "genuineness": 4.18726921081543, "identity": 0.5815910696983337, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/e3ebf0cf23da140cc3ca6986f0c51012966f6b02bca1ac151974bbaa479dd30a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "e50c55f4a1b2a076343fd2928fcbb7f5d244c7aae29792467aa0e5f6689d531f", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "d9fe202c6eb76ec8437cc3da6933f0f35f56e51c1c31b640d37742a57eb0f8d6", "generation_seconds_batch": 16.632900099968538, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9363051652908325, "target_z": -1.4218750364967323, "ce": 6.252951145172119, "genuineness": 4.457139015197754, "identity": 0.6837610006332397, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/e50c55f4a1b2a076343fd2928fcbb7f5d244c7aae29792467aa0e5f6689d531f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "e5d516b83e1d174c853ea606458536942224f3864d77834ff019505dd020ed6e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ab4408603232e962f0c0f4f456f26b40aed4410fb06301622f4cd0d28e127929", "generation_seconds_batch": 25.850319700082764, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.433278560638428, "target_z": 0.5351562218975161, "ce": 5.784333229064941, "genuineness": 4.7081804275512695, "identity": 0.6888359189033508, "hypothesis": "He encontrado la cajita, estaba junto la ventana vieja. ¿podemos mirar juntos lo que hay dentro?", "wer": 0.058823529411764705, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/e5d516b83e1d174c853ea606458536942224f3864d77834ff019505dd020ed6e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "e5f6c08f8081fd105a2e4572996f10b43b1c3884432fa20327126f2d38d2b0c4", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f45fa22eec73ef4899b2eeb4932835b1e43a8e192ff2921d962ae30d6e6de2ab", "generation_seconds_batch": 22.606799453962594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7229058742523193, "target_z": -0.02160643923351303, "ce": 6.125868320465088, "genuineness": 4.607763767242432, "identity": 0.7357873320579529, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que entró?", "wer": 0.11764705882352941, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/e5f6c08f8081fd105a2e4572996f10b43b1c3884432fa20327126f2d38d2b0c4.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "e627e011ba41c1a566b9c268eac1bd28bd4c84585db8dcf6a1b321414bcdfd82", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "0cfd17da8afbb75db9c4d78628d2d984101818d5abed6a3be083486b4008b990", "generation_seconds_batch": 23.053280350985005, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5261945724487305, "target_z": -0.17578115620339804, "ce": 6.117925643920898, "genuineness": 4.457139015197754, "identity": 0.7284865379333496, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/e627e011ba41c1a566b9c268eac1bd28bd4c84585db8dcf6a1b321414bcdfd82.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "e66a81187f5400a108c4d00cbdf4a128bcb18c7b7d68f8edeced9e2a12ddbb94", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "1eed9fe5f5dcf9c53d84746c30fae2b2a3e58094e8b339156b8559debe2d8094", "generation_seconds_batch": 22.28225635702256, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9722604751586914, "target_z": 0.1738280613132022, "ce": 5.831989288330078, "genuineness": 4.457139015197754, "identity": 0.6835607886314392, "hypothesis": "De deur staat nog open. Ik heb je het jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/e66a81187f5400a108c4d00cbdf4a128bcb18c7b7d68f8edeced9e2a12ddbb94.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "419e0b4a9317c50bf75329590a8533415d39b7f99efeb5d3ce5a8afc1abf0c72", "generation_seconds_batch": 17.287815985037014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5922322273254395, "target_z": -0.12402339416013042, "ce": 6.459460258483887, "genuineness": 4.469690799713135, "identity": 0.4090225100517273, "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_CASU_reg/e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "e7000ba720cc77591458fcb3c7909834020ad7d4f348a17d0af92fd8dff10b98", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c9a22700408131a54d5c1aca248bd0a15747393aa4093a07b6492f321e9926da", "generation_seconds_batch": 23.29418758803513, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.248871326446533, "target_z": 0.3906249576637906, "ce": 5.9431867599487305, "genuineness": 4.362998008728027, "identity": 0.5068529844284058, "hypothesis": "De deur 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_CASU_reg/e7000ba720cc77591458fcb3c7909834020ad7d4f348a17d0af92fd8dff10b98.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "e7be6b235e4370d12383996d1949cab7712c12b97aa3ed6a7ef808750b8e7a20", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "b81cc3f535439dead5f85557163dd4f947d701ca1971093b1c281b2a3204ea5d", "generation_seconds_batch": 15.347349507035688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.104335784912109, "target_z": 0.2773435853997374, "ce": 6.070269584655762, "genuineness": 4.369274139404297, "identity": 0.624329686164856, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/e7be6b235e4370d12383996d1949cab7712c12b97aa3ed6a7ef808750b8e7a20.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "e8a7fb46ba6e29c1b2aa0982b1181e3892340cb50ca3c836d6a9ddc7f952bd54", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "14f3e9c970aac7c00516284900502f7c5e2dc31d1432d6aefa9b945a98680877", "generation_seconds_batch": 16.66787780507002, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.255280017852783, "target_z": -1.1718748729913717, "ce": 5.95112943649292, "genuineness": 4.243753433227539, "identity": 0.49459022283554077, "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_CASU_reg/e8a7fb46ba6e29c1b2aa0982b1181e3892340cb50ca3c836d6a9ddc7f952bd54.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": "e8be3f0533a86934ef5ba0d69c2e3ed2a871241ceb4957a57172620f1e2a165d", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe3522a041660f92152553a8325829e112e987d75ca127684578dc486db67b8d", "generation_seconds_batch": 16.747511810041033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.029576301574707, "target_z": 0.21874999124078426, "ce": 5.863759994506836, "genuineness": 4.6956281661987305, "identity": 0.7643799185752869, "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_CASU_reg/e8be3f0533a86934ef5ba0d69c2e3ed2a871241ceb4957a57172620f1e2a165d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "e8ddb8d1dc39be3b14128aeb19e37154abf9d846811973ad84001e7a196f3194", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "aa4c632e07145023430e9ae600b61c70685b3664eeb049035d50a7e6f4430fcb", "generation_seconds_batch": 23.46913131303154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.414767265319824, "target_z": -1.0468749313861434, "ce": 6.3641486167907715, "genuineness": 4.595211505889893, "identity": 0.49215707182884216, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/e8ddb8d1dc39be3b14128aeb19e37154abf9d846811973ad84001e7a196f3194.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": "e8ea713784b83d45695ae6f31aabc81a92ae8757d54e268e92ebfeb9d581b32b", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0084d77c1510adb43e132538b9f3e82e3bafd88358b26fbad17beb1d84dedfd9", "generation_seconds_batch": 17.089458236005157, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.003410339355469, "target_z": 0.1982421211671891, "ce": 5.824047088623047, "genuineness": 4.607763767242432, "identity": 0.7330693006515503, "hypothesis": "La route tournait à gauche après le pont, continue jusqu'à la petite barrière en bois.", "wer": 0.3, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/e8ea713784b83d45695ae6f31aabc81a92ae8757d54e268e92ebfeb9d581b32b.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": "e8efadefdb641acddb86fdee0e9d7b32892734c81966ae43ded47f411bf8b02d", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "26f8a449ac866b943f578b4e2b02a797b79f0a1289f9ca137f09577d29c25ea1", "generation_seconds_batch": 24.549607152934186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6022000312805176, "target_z": -0.11621101459934693, "ce": 5.95112943649292, "genuineness": 4.545003414154053, "identity": 0.6129870414733887, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zu dem kleinen Holztor. Ich fahr auf der anderen Seite.", "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_CASU_reg/e8efadefdb641acddb86fdee0e9d7b32892734c81966ae43ded47f411bf8b02d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "e8f3f3b07a09056af77eaa35bf2afd817cdd44b35a7491e18fb215d2188aa532", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5f25ac2b4e246366016277aad3e8037dcc903afb6d351843b373cb67146c485", "generation_seconds_batch": 21.620567710953765, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5810182094573975, "target_z": -0.1328125080292811, "ce": 6.181467056274414, "genuineness": 4.632867813110352, "identity": 0.597435474395752, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/e8f3f3b07a09056af77eaa35bf2afd817cdd44b35a7491e18fb215d2188aa532.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": "e98dfb1baaa5dd678abe182a017a8cfb82b5e753dd67cf21a66ae8d10c23f3fe", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "eafec4a49f7da021ff2a40b6d58a7a1bc478423c746e7449e75556fd234b7908", "generation_seconds_batch": 22.77398750500288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9348807334899902, "target_z": 0.1445312642337256, "ce": 5.601652145385742, "genuineness": 4.482243061065674, "identity": 0.6539306640625, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/e98dfb1baaa5dd678abe182a017a8cfb82b5e753dd67cf21a66ae8d10c23f3fe.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "e9966a6d9a1e444e7d6ea8979ecf7efe1d67153884a3a6b2880da9fe9cf9148d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1949f62d097bc35f76a431c3c75bc6969a10f48b3866b6430d959217de6670c7", "generation_seconds_batch": 21.714354204945266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8819260597229004, "target_z": 0.10302743722725553, "ce": 5.927301406860352, "genuineness": 4.65797233581543, "identity": 0.16934829950332642, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/e9966a6d9a1e444e7d6ea8979ecf7efe1d67153884a3a6b2880da9fe9cf9148d.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "f9072a7f0094df6d0a8e487152a93f55e884432e185a5363c7f3aea896094fda", "generation_seconds_batch": 19.275168676045723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.095792531967163, "target_z": -1.2968750014598693, "ce": 6.4515180587768555, "genuineness": 4.469690799713135, "identity": 0.502685010433197, "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_CASU_reg/e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41.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": "ea4a526ea1d3345809a2fd9d93c4e12f18e6ef5babb1a4b715ba9b09dde24e1a", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Someone moved the little blue chair.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a073629ff268164075863bb2736b0539fd32ee4fe54e9e7df7b61755b4ae956d", "generation_seconds_batch": 24.254928041016683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.248871326446533, "target_z": 0.3906249576637906, "ce": 6.23706579208374, "genuineness": 4.557555198669434, "identity": 0.33385586738586426, "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_CASU_reg/ea4a526ea1d3345809a2fd9d93c4e12f18e6ef5babb1a4b715ba9b09dde24e1a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "ea79e1a887dda277f64f746029956e5701d11e70bf0fca5973da2c38b3ce8345", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "f93c6988a8cc729a509f91b846e884aa08a813a32affad19f1dfcbce4aba9a6a", "generation_seconds_batch": 25.129749748972245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.324343204498291, "target_z": -0.33398445839503327, "ce": 6.221180438995361, "genuineness": 4.419482707977295, "identity": 0.7767196893692017, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ea79e1a887dda277f64f746029956e5701d11e70bf0fca5973da2c38b3ce8345.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "eae0245935cf8d00e3c26e67e443d3c74134febcfff4021da7f21d30742afbe6", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "b4c5401cbb8e482fdb8bf9a5817107cc0a5dc41cd27a6f7e34f0de90bc4ae00b", "generation_seconds_batch": 23.177522643003613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.08262038230896, "target_z": -0.5234374656930717, "ce": 6.24500846862793, "genuineness": 4.65797233581543, "identity": 0.6694532632827759, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/eae0245935cf8d00e3c26e67e443d3c74134febcfff4021da7f21d30742afbe6.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "eb0a618b0c95520cf09cacb7ab6df15efb153747d8666a5ac55127c9b53835b6", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "6f60240916bd2a1d94c83ac261a8249ee5cd4139e1b5d7cde8d6a3c3e7874d2b", "generation_seconds_batch": 22.312673361971974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3150875568389893, "target_z": -1.125000035036863, "ce": 6.475345611572266, "genuineness": 4.406930446624756, "identity": 0.7168956995010376, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/eb0a618b0c95520cf09cacb7ab6df15efb153747d8666a5ac55127c9b53835b6.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": "eb592ad63851660459c218ddccf1229b3dcc85546efe1a40d76a49a7cf56c466", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Your message arrived after I had left.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "639307c5613dc0cb05c0841ff28f17b534a65e6d1e50ef727dd577ab21a622de", "generation_seconds_batch": 21.428442695993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6183979511260986, "target_z": -0.10351571094980451, "ce": 6.284721851348877, "genuineness": 4.557555198669434, "identity": 0.4348244369029999, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/eb592ad63851660459c218ddccf1229b3dcc85546efe1a40d76a49a7cf56c466.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "eb83a5f19057cf1e7034cc6f5114c76c5e8a8c29af33dd42b224b1b1ba096636", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "c5affb0d2f9c3f594c9c49d311d989d76420618efc157029ef7b422c07428a65", "generation_seconds_batch": 21.78946642507799, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.797109127044678, "target_z": 0.8203125605845756, "ce": 6.141753673553467, "genuineness": 4.632867813110352, "identity": 0.3576280474662781, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/eb83a5f19057cf1e7034cc6f5114c76c5e8a8c29af33dd42b224b1b1ba096636.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "7db3d0352e0f5729cbc0656d1038217263a7b723eab5ea1c566f896abeb641eb", "generation_seconds_batch": 17.592919258982874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.559302568435669, "target_z": -0.9335937459853595, "ce": 6.149696350097656, "genuineness": 4.406930446624756, "identity": 0.7240064740180969, "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_CASU_reg/ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "ebef001d93cb64e13c4f6fc280434f9217a3e1cfee936d4a446b769a52f7794e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "920d1abf016cdd5ef86741d65951b4464ab32df92f12ca1d4f341e40318d18f7", "generation_seconds_batch": 22.008470964967273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7967537641525269, "target_z": -1.5312500321171243, "ce": 6.427689552307129, "genuineness": 4.086852550506592, "identity": 0.5127853751182556, "hypothesis": "Ich habe die kleine Schacht gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ebef001d93cb64e13c4f6fc280434f9217a3e1cfee936d4a446b769a52f7794e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "ec258bc6188e3921d6ff8bf1aee8b417465a81fd45e8b7663fa30d3996fa9446", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c68598ef02b3df2cf3b300500450a9dadbd83a3fd39af75f42053592d18d9110", "generation_seconds_batch": 24.354885635897517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4040870666503906, "target_z": -0.2714843007291498, "ce": 5.522225379943848, "genuineness": 4.570107460021973, "identity": 0.7695754170417786, "hypothesis": "J'ai trouvé la 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_CASU_reg/ec258bc6188e3921d6ff8bf1aee8b417465a81fd45e8b7663fa30d3996fa9446.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "ec650c20272b02839deba157b3b076b2321caf9518dd02e016b0939037fc958a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "e88f65f694f6dc45758c8d17336d99961556c6dac03f5eb57224f98d9676ea7f", "generation_seconds_batch": 24.17917200003285, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8465936183929443, "target_z": -1.4921874802917645, "ce": 5.895530700683594, "genuineness": 4.068024158477783, "identity": 0.7074847221374512, "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_CASU_reg/ec650c20272b02839deba157b3b076b2321caf9518dd02e016b0939037fc958a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "ec6b130dfaae414bc5a4ae7c959cca7f6c89abe8d49ff9672562f7b90cef860c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "39f12fd1ac819cb0eab06b196e7fcf70cc93ca0c22a0f02d43769d9a99b71d28", "generation_seconds_batch": 22.687641764990985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2353439331054688, "target_z": -1.1875000058394771, "ce": 6.197352409362793, "genuineness": 4.093128204345703, "identity": 0.42882102727890015, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ec6b130dfaae414bc5a4ae7c959cca7f6c89abe8d49ff9672562f7b90cef860c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "ecd4f3ee46478aaf42f3e8dae6ac16f0805eff9256426c794b12a89ea8964b96", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "f80fa01bb2c77cb0b62f054952f32bc897086ca446592cacfd85fef045ccb6f5", "generation_seconds_batch": 16.90145147300791, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.199031829833984, "target_z": 0.3515626861333346, "ce": 6.157639026641846, "genuineness": 4.607763767242432, "identity": 0.3527975082397461, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ecd4f3ee46478aaf42f3e8dae6ac16f0805eff9256426c794b12a89ea8964b96.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "ecf4f7d8aceac6e1d5524ede717a29d37dacd212964a40809cb81bae8286a3bd", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "3529f26a23c63889a69c4ef1676e7ef002881de2050eff12e23efd0451b0b922", "generation_seconds_batch": 16.27507096901536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.095792531967163, "target_z": -1.2968750014598693, "ce": 6.419747352600098, "genuineness": 4.344170093536377, "identity": 0.5726867914199829, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ecf4f7d8aceac6e1d5524ede717a29d37dacd212964a40809cb81bae8286a3bd.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "eda0542566b2d6a076f9832d45861709e1832f1865194b385b9bf1312c94959d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "0f648151c22186a768eb3f04151e7e336d1f02021086748ace7dbb8d464d372a", "generation_seconds_batch": 22.793955393019132, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8932290077209473, "target_z": -0.6718751065704582, "ce": 6.252951145172119, "genuineness": 4.645420074462891, "identity": 0.603420615196228, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/eda0542566b2d6a076f9832d45861709e1832f1865194b385b9bf1312c94959d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "ee32a0444e2309d8eac93cd8775726d3ce4ab3218a4ce9f36a2457b58225f5f4", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "9b2f8f73072efb5656a2f49e73f39516cb69fe8eba3d00077133c83a418d1e42", "generation_seconds_batch": 23.800322041963227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.228935241699219, "target_z": 0.37499982481568506, "ce": 6.348263263702393, "genuineness": 4.607763767242432, "identity": 0.4217489957809448, "hypothesis": "I f found the 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_CASU_reg/ee32a0444e2309d8eac93cd8775726d3ce4ab3218a4ce9f36a2457b58225f5f4.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": "ee58b809df341da2396fcdd0b33e281c64fb0dcb913e927809a572833168cf98", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Someone moved the little blue chair.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9166ae1c7d83565ac1268f95e0e8f4fdfea8a65c096c015855472b47519879eb", "generation_seconds_batch": 24.254928041016683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.588494300842285, "target_z": -0.12695303649542422, "ce": 6.109982967376709, "genuineness": 4.494794845581055, "identity": 0.26005518436431885, "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_CASU_reg/ee58b809df341da2396fcdd0b33e281c64fb0dcb913e927809a572833168cf98.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "ee7edf3a1ec1888ca4a661f46c73ef1b942b5a645ce36c03a9e057977f7351c1", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "13ffd805ba67cedafb3284c07647a64a7d7b99cfbaa55e5204740ef4ecdc6763", "generation_seconds_batch": 17.171801276039332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.969768524169922, "target_z": 0.17187496642300631, "ce": 6.006728172302246, "genuineness": 4.369274139404297, "identity": 0.36087697744369507, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ee7edf3a1ec1888ca4a661f46c73ef1b942b5a645ce36c03a9e057977f7351c1.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "ef38b7e42d1cd44bdcbbde96695226f76036f8336652c7cf516d02a1fe695fd9", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cd5d239c61ebc395e34a3eba484c6ba63dfefcd3f6b350e1b97da5ec09f224eb", "generation_seconds_batch": 17.193021383020096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.169127941131592, "target_z": 0.3281251737244456, "ce": 5.748591423034668, "genuineness": 4.6203155517578125, "identity": 0.7079982757568359, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ef38b7e42d1cd44bdcbbde96695226f76036f8336652c7cf516d02a1fe695fd9.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "efb4bcc814fe7196707572db59b861a718e1a983f2d6b5e914fc6af3fbcceda0", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "ff1b0d519dc34ba86de291262937d4d5bc5b7e13d66cd7e3aae7e28c8e8961f3", "generation_seconds_batch": 24.424999363021925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.077636241912842, "target_z": -0.5273438423367327, "ce": 6.133810997009277, "genuineness": 4.256305694580078, "identity": 0.6007436513900757, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/efb4bcc814fe7196707572db59b861a718e1a983f2d6b5e914fc6af3fbcceda0.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": "f019667b0044a6b2650808386dce0c7c94fbef48c0c1f44f5d15f606386392c0", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "15d58cda242b3ba5aff95d25edbe4604b3664f84f343ecdb05e44dca8d734aef", "generation_seconds_batch": 16.14351812598761, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.461402654647827, "target_z": -0.226562557664837, "ce": 6.038498878479004, "genuineness": 4.557555198669434, "identity": 0.8299875259399414, "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_CASU_reg/f019667b0044a6b2650808386dce0c7c94fbef48c0c1f44f5d15f606386392c0.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-1", "key": "f0231982a5586e1341da326f7cda1c1ff144cad886f7178f1f3fc0ec4a91448c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond naast het oude raam.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "0328c4a60655bfffa2ddef5d0cb909823e7305a8ae6a5bdc5933faecda32b957", "generation_seconds_batch": 25.158052872982807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.779754400253296, "target_z": 0.022949238686340006, "ce": 6.141753673553467, "genuineness": 4.444586753845215, "identity": 0.5131851434707642, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/f0231982a5586e1341da326f7cda1c1ff144cad886f7178f1f3fc0ec4a91448c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "f0274fa66378f03c855c65cdfbda6d01a22ce4a2437096dc4649f18e5b1aa085", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "d6dee368eb3f9642a519d4314344aa52348f324efd54c9a6800103e3fa28eda6", "generation_seconds_batch": 22.65880296996329, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.513022422790527, "target_z": 0.5976563795633996, "ce": 5.990842819213867, "genuineness": 4.218649387359619, "identity": 0.3636358082294464, "hypothesis": "De deur staat nog open. Ik heb je Jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/f0274fa66378f03c855c65cdfbda6d01a22ce4a2437096dc4649f18e5b1aa085.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "f0bc58b1de44ac7a11626f470bd4b53575ac791fc7073c9079eb55f099a87ee4", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a6638f05bd31a930fdaf01512fc0e7ff83ca62826dc5ab0a6d7900d442cf50f8", "generation_seconds_batch": 16.155976719921455, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.104335784912109, "target_z": 0.2773435853997374, "ce": 5.9431867599487305, "genuineness": 4.482243061065674, "identity": 0.16552600264549255, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/f0bc58b1de44ac7a11626f470bd4b53575ac791fc7073c9079eb55f099a87ee4.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "f0dd07e95637b0114fa733f0f8671172fada2c5ca8df2d129b8c24772cc5b4a7", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "781dab3e9b221bcdc70337499dbe60e71031194fc155f33d10a9e6aa68dd7026", "generation_seconds_batch": 22.701795726898126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.195472240447998, "target_z": -1.2187498978091496, "ce": 6.181467056274414, "genuineness": 4.306513786315918, "identity": 0.7133626937866211, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/f0dd07e95637b0114fa733f0f8671172fada2c5ca8df2d129b8c24772cc5b4a7.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "d06cdcc2cae74dc8a642c414241b244ebf4b722f7f5af42d417b29666c6923c2", "generation_seconds_batch": 18.585101302014664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8533573150634766, "target_z": -0.7031249985401307, "ce": 6.252951145172119, "genuineness": 4.381826400756836, "identity": 0.7672373056411743, "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_CASU_reg/f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8.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": "f22298ceb28f7ca3ed09a5a5bb7a998037037a2b3e2d411b436f7df1d6462e5c", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The road turns left behind the bridge.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f64d1b1bdb6764472dc4de6f89059ea263a11882c5c1a110a35d673e4a16030", "generation_seconds_batch": 17.88598236395046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9929087162017822, "target_z": -0.5937500029197386, "ce": 6.316492557525635, "genuineness": 4.532451152801514, "identity": 0.39207470417022705, "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_CASU_reg/f22298ceb28f7ca3ed09a5a5bb7a998037037a2b3e2d411b436f7df1d6462e5c.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": "f23ffe2e8187938778f958c112bc760925143b17e48817d64c7e2e0ae7a8da5d", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dcd67cbc4ba4d3645ae7a80e23ac7c395cbf7ed13e7d99d04c9a6b218e74583c", "generation_seconds_batch": 16.881773566943593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.617685794830322, "target_z": 0.6796874861312417, "ce": 6.070269584655762, "genuineness": 4.457139015197754, "identity": 0.32959887385368347, "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_CASU_reg/f23ffe2e8187938778f958c112bc760925143b17e48817d64c7e2e0ae7a8da5d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "f387f0cdf5133f3b1f57e4188d91f62130f97861977e2c74a57fef98a1ff8eac", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "955afc42c9a9d55ba0d09877cdb9ee50ed736573d63153521a0ef3fd6be7c5a6", "generation_seconds_batch": 23.110295646009035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7006337642669678, "target_z": -0.039062458393725207, "ce": 5.613565921783447, "genuineness": 4.595211505889893, "identity": 0.4867088198661804, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/f387f0cdf5133f3b1f57e4188d91f62130f97861977e2c74a57fef98a1ff8eac.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "f3a8a394efab06088ae173732f9264173a80460f7b804c9f5971833d7245c5ba", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "ab91938a6e446879003ebd9a478c3d2743869488bf719ad5450f54b604fabda8", "generation_seconds_batch": 24.664431971963495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.149191856384277, "target_z": 0.31250004087634015, "ce": 5.474569320678711, "genuineness": 4.833701133728027, "identity": 0.7150423526763916, "hypothesis": "He encontrado la cajita. Estaba junto la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/f3a8a394efab06088ae173732f9264173a80460f7b804c9f5971833d7245c5ba.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "f3c1bb0acea8ee50c50bab593976b3e744c525be61a4644f16a267a9ee80b98a", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ba419c5648a88ac87068ddda8463d8254f46716c82b0e706fd567c5dc4ce11ab", "generation_seconds_batch": 24.606192229082808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.333599090576172, "target_z": 0.45703130511006573, "ce": 5.689021587371826, "genuineness": 4.6203155517578125, "identity": 0.631050169467926, "hypothesis": "La porte est encore ouverte j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/f3c1bb0acea8ee50c50bab593976b3e744c525be61a4644f16a267a9ee80b98a.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "f3fe00fceb621e8c3155b6eca8b72635dcb9d48afc286b88e2860378fd816ef9", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "4b05b99776b03faed57b10366d57a621a988b3738706ce45600c2368d3736782", "generation_seconds_batch": 25.063222842058167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2171878814697266, "target_z": -0.4179686598530713, "ce": 6.252951145172119, "genuineness": 4.331617832183838, "identity": 0.1579119861125946, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/f3fe00fceb621e8c3155b6eca8b72635dcb9d48afc286b88e2860378fd816ef9.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": "f56af92f852e1f45eced8da12affd7c599b22c97d3c4059f4aae2058842de6d8", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ec5beec9d017b90096d115db37297ce31caf00f688aa9dd4720dc2463c176816", "generation_seconds_batch": 17.667625197093002, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9598007202148438, "target_z": 0.1640625868622228, "ce": 6.062326908111572, "genuineness": 4.532451152801514, "identity": 0.548377275466919, "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_CASU_reg/f56af92f852e1f45eced8da12affd7c599b22c97d3c4059f4aae2058842de6d8.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "f59a34ec37789dcb37577f8656494a263ac4ee595a940d72448f25ee8c1d82b6", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "96bd31659cc476570ad5cb389aa214960b132a6b2005fdfa93a6290ee4ece5b3", "generation_seconds_batch": 19.78335841803346, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9872124195098877, "target_z": 0.18554681751764668, "ce": 6.213237762451172, "genuineness": 4.65797233581543, "identity": 0.5739794373512268, "hypothesis": "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_CASU_reg/f59a34ec37789dcb37577f8656494a263ac4ee595a940d72448f25ee8c1d82b6.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "f59bc9ac10c1ab5bb2a03b7c993e5df2405a21a3d184d27ece68b6e32bcef606", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a54806334fd05964e2657a228e732f32e1ed6cdd09dd8e1c27f252b04e936591", "generation_seconds_batch": 23.965668428922072, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3991029262542725, "target_z": -0.2753906773728108, "ce": 5.808161735534668, "genuineness": 4.425758361816406, "identity": 0.7838071584701538, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/f59bc9ac10c1ab5bb2a03b7c993e5df2405a21a3d184d27ece68b6e32bcef606.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "f59f158750256e88d9fc0ed81ec6cf5fa2bec960bf2bd06a910eb80034d82cf5", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aa25511d40993c5c73ba4505a9b05c5689d1413ca37bbc28cfaa245a090f630f", "generation_seconds_batch": 16.657624796964228, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.673933982849121, "target_z": -0.8437500729934646, "ce": 5.879645347595215, "genuineness": 4.482243061065674, "identity": 0.7717890739440918, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/f59f158750256e88d9fc0ed81ec6cf5fa2bec960bf2bd06a910eb80034d82cf5.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "f68b36e56a61e57c8c38cd160216824c7689ffc79863db5102b9b4014fa9f1ae", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "1f5c8e67307f3e6d2edbc60431cb96cf36bd67b5bbbb880d693529b9fce76c25", "generation_seconds_batch": 21.289421800989658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.92313289642334, "target_z": -0.6484375941615693, "ce": 6.0305562019348145, "genuineness": 4.381826400756836, "identity": 0.4685497283935547, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/f68b36e56a61e57c8c38cd160216824c7689ffc79863db5102b9b4014fa9f1ae.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "f6e2780904bd3007ef77018a1de9ddd1deaaf0d4085782b3b701a39d3d72e330", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "02fc7bae0490038b79a7caad1cdb76d4c93a7bdd10dc2f82385f093ac9ca8ff0", "generation_seconds_batch": 16.74104550492484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.458198547363281, "target_z": 0.5546875445260133, "ce": 5.879645347595215, "genuineness": 4.65797233581543, "identity": 0.708421528339386, "hypothesis": "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_CASU_reg/f6e2780904bd3007ef77018a1de9ddd1deaaf0d4085782b3b701a39d3d72e330.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "f6ebdc84f2b68661707494f2a913603e063ff8866e579c89f1c4d6c6146e0d72", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "a48a145b03f9f62a02dac1c5e8c6939609c6434165c108cdc1b047d197d72f29", "generation_seconds_batch": 24.34107729594689, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.283759117126465, "target_z": 0.4179686598530713, "ce": 5.895530700683594, "genuineness": 4.68307638168335, "identity": 0.8157795667648315, "hypothesis": "La porte est encore ouverte j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/f6ebdc84f2b68661707494f2a913603e063ff8866e579c89f1c4d6c6146e0d72.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "f707364533d10d23a170c5d072cdfaa3953fe931add33e3d99ac79512f1fd290", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "96cbf3315836f020f87e202da9b1f0a4aa136ea322aaea879f2391c7def542d2", "generation_seconds_batch": 22.620094978949055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5443506240844727, "target_z": -0.9453125021898039, "ce": 6.022613525390625, "genuineness": 4.005263805389404, "identity": 0.655978798866272, "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_CASU_reg/f707364533d10d23a170c5d072cdfaa3953fe931add33e3d99ac79512f1fd290.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "636bf23f6a0452bf5d2f855673688aef30df343a175ec993bff064a7fb0cd233", "generation_seconds_batch": 16.508388701011427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7935495376586914, "target_z": -0.7500000233579086, "ce": 6.229123115539551, "genuineness": 4.432034492492676, "identity": 0.7324117422103882, "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_CASU_reg/f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "f80b4dcb330ea408adb96f47128d7335decae4859a0cc3fec9b725566068881d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "0f6e6104d5316dada320e8639b228b1d9bc2796b427d39d85b3e675ac27e075d", "generation_seconds_batch": 25.101643697009422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5941903591156006, "target_z": -0.9062500437960788, "ce": 6.221180438995361, "genuineness": 4.507347106933594, "identity": 0.4576345682144165, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/f80b4dcb330ea408adb96f47128d7335decae4859a0cc3fec9b725566068881d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "f84bfa6d8708e3c17a8ace945243e8655a91eb8fd8edf7f4f0bb12ff49bc8b94", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "07d5425b6e979ebd06bc19a48e969484b5dd8db79b89fce13d6096aced8934cf", "generation_seconds_batch": 21.84295719896909, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.468166351318359, "target_z": 0.5624999240867968, "ce": 5.190618991851807, "genuineness": 4.570107460021973, "identity": 0.7195354700088501, "hypothesis": "La puerta sigue abierta. ¿Hejado tu abrigo sobre la silla? Dime cuándo estás listo.", "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_CASU_reg/f84bfa6d8708e3c17a8ace945243e8655a91eb8fd8edf7f4f0bb12ff49bc8b94.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_CASU_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": "d821eb8b1037d81d42284b0e3231b139c373533128cd708aebfed1a6b4fe4956", "generation_seconds_batch": 18.3169145709835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5343828201293945, "target_z": -0.9531248817505874, "ce": 6.300607204437256, "genuineness": 4.432034492492676, "identity": 0.562355101108551, "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_CASU_reg/f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "f8c2b3e45fabdc90ffa7493d116522e4a1c57ac627ef7bb2e7f1ceb343563e73", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "c3f400c5209b0938b30cda7aef4107f7ff3dad50b738c01b4bec7d4a537c3666", "generation_seconds_batch": 22.68132184201386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.861900806427002, "target_z": 0.8710937751827452, "ce": 5.8875885009765625, "genuineness": 4.2751336097717285, "identity": 0.5331606268882751, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/f8c2b3e45fabdc90ffa7493d116522e4a1c57ac627ef7bb2e7f1ceb343563e73.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "f9577e2639cf3cd13695cc0fe94e8ca154b0e9880628e0c2fc3e6b2e91efadbc", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b99aa93f5c50ee1f3e4f33a3c78b8312016a257332a93c36df1f2ffec4d1e17", "generation_seconds_batch": 22.73331608402077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.281979560852051, "target_z": -0.3671874452549016, "ce": 5.871703147888184, "genuineness": 4.457139015197754, "identity": 0.6633267402648926, "hypothesis": "La porte est encore ouverte et j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/f9577e2639cf3cd13695cc0fe94e8ca154b0e9880628e0c2fc3e6b2e91efadbc.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "f9709dd9a2163a9881a25ce0afe00185bfe5a89626ecdcf0f4406b5d107755b1", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "efb8761b20f9428456d3a17f56c488ab2abf8c8529b204e79238d9e8de6c4cc2", "generation_seconds_batch": 23.706792637938634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1125242710113525, "target_z": -0.4999999532841827, "ce": 5.974957466125488, "genuineness": 4.306513786315918, "identity": 0.34031274914741516, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/f9709dd9a2163a9881a25ce0afe00185bfe5a89626ecdcf0f4406b5d107755b1.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "f99e696620cd64739f029dbce5243a403f8ca9c1f7ff97862b4d688a5a638a1b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "f0c51657c564a0ddd26a56a8e75b19e24ee96c2289d787bb7b30dd5acb4a9861", "generation_seconds_batch": 26.80855601897929, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.973506450653076, "target_z": 0.17480460875830012, "ce": 6.348263263702393, "genuineness": 4.557555198669434, "identity": 0.207992285490036, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/f99e696620cd64739f029dbce5243a403f8ca9c1f7ff97862b4d688a5a638a1b.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "fa5ce58b88f989e207b367bc7f32312920186f3fff433c7b21b44a28dc092add", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "8f49a8f3cf8498d8f6715a1ea824ac2bcc78934802c43dd8a737a552195b40ec", "generation_seconds_batch": 16.921104872017168, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6264970302581787, "target_z": -0.09716796569339867, "ce": 5.903473377227783, "genuineness": 4.7081804275512695, "identity": 0.8022825717926025, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/fa5ce58b88f989e207b367bc7f32312920186f3fff433c7b21b44a28dc092add.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "fa5d93715f832a02b8837cb90f2f2e61ce30f942a959fa8bec690c5d82f4d496", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "74f2e59da90355983dfd76b0e4262e17a3a72a031d62bf2fb7aa4268acc4c870", "generation_seconds_batch": 21.825582813005894, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7884762287139893, "target_z": 0.02978507080202556, "ce": 5.77639102935791, "genuineness": 4.632867813110352, "identity": 0.7132251262664795, "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_CASU_reg/fa5d93715f832a02b8837cb90f2f2e61ce30f942a959fa8bec690c5d82f4d496.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "fa839a61042aa324c1a492c3397e143a7f02015a9c89abc1e21fc5d3e2017b12", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "4cee6012e756c4ecec32cf38574bddf6336f62f5f62410c0fa6189d5d143439e", "generation_seconds_batch": 22.763790930970572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.028330326080322, "target_z": 0.2177734437956863, "ce": 6.260893821716309, "genuineness": 4.519899368286133, "identity": 0.22422778606414795, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/fa839a61042aa324c1a492c3397e143a7f02015a9c89abc1e21fc5d3e2017b12.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "fa963f6509a1812fa55c3a2b3c2ec357adec2291ec592106c95cc1e30ee652ea", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "28b10b4a563e294d361d8c8a473a881168d0834936640fc346055207cff080ef", "generation_seconds_batch": 23.137088172021322, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5187184810638428, "target_z": -0.18164062773725492, "ce": 6.221180438995361, "genuineness": 4.65797233581543, "identity": 0.7387874126434326, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/fa963f6509a1812fa55c3a2b3c2ec357adec2291ec592106c95cc1e30ee652ea.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": "faa36164036dfd71bf1c063abaf5a24010f280b9ba5b883abd51b8cac05a5a65", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The keys are on the kitchen table.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d51f2e8e15dd724c51e8f06583166f8b8cd8eced4c541af7fe48959e2ae7078", "generation_seconds_batch": 17.144701304961927, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6775827407836914, "target_z": -0.05712895985457553, "ce": 6.268836498260498, "genuineness": 4.381826400756836, "identity": 0.5434409976005554, "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_CASU_reg/faa36164036dfd71bf1c063abaf5a24010f280b9ba5b883abd51b8cac05a5a65.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "fab1700d3e7ce80cbeb11049d1741d2d438bc8ca165d29b6d66b4cd7748e78af", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "85bf68771941042f7d38920cdd79fdcd039f6deabbf6d3224f5dfbcae82a4ca9", "generation_seconds_batch": 17.38991058699321, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.114304065704346, "target_z": 0.2851563386870594, "ce": 5.871703147888184, "genuineness": 4.7332844734191895, "identity": 0.8061457872390747, "hypothesis": "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_CASU_reg/fab1700d3e7ce80cbeb11049d1741d2d438bc8ca165d29b6d66b4cd7748e78af.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "fb3ccd9f5243c70cb51a5eaf44ec6743d212756cc540435d65a9c1c24fdbb5e7", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "bdf1b91338368685f08d127bd0f9559ccde69fb02c52dcefc82d0fcc9ed73327", "generation_seconds_batch": 21.999566774000414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.673933982849121, "target_z": -0.8437500729934646, "ce": 5.990842819213867, "genuineness": 4.406930446624756, "identity": 0.45531919598579407, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/fb3ccd9f5243c70cb51a5eaf44ec6743d212756cc540435d65a9c1c24fdbb5e7.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-1", "key": "fb73fd944b1a35514785a7c44908d9aab979e7b0dbfe2df17a0778d4a5a34b5d", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "b4de18ae7a087d14e2d5322584d3baf14a4cbb27ea430308536af79ff3838046", "generation_seconds_batch": 25.470492660999298, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5973947048187256, "target_z": -1.68749995912366, "ce": 6.260893821716309, "genuineness": 4.400654315948486, "identity": 0.6989283561706543, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/fb73fd944b1a35514785a7c44908d9aab979e7b0dbfe2df17a0778d4a5a34b5d.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": "fbfc98e3807d3a573c7ca9880260d07cd7da26b755119996101df40d9f917c8b", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Escribí el número en el sobre.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "311607af267dfa52204294e9e4e08103474dccc0349f3d5075b35e678993b88c", "generation_seconds_batch": 15.098701035953127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6928462982177734, "target_z": -0.04516597335722193, "ce": 5.824047088623047, "genuineness": 4.808597087860107, "identity": 0.6732439994812012, "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_CASU_reg/fbfc98e3807d3a573c7ca9880260d07cd7da26b755119996101df40d9f917c8b.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": "fc34fd97788acf1cc690744e002cad917889791f9ee1230f73e38aac10eb6750", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "72319dc77d04af8565f890867478a42cf7bf24f440d38d74024155db78c8131c", "generation_seconds_batch": 25.09243169601541, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.228935241699219, "target_z": 0.37499982481568506, "ce": 6.09409761428833, "genuineness": 4.369274139404297, "identity": 0.6414403915405273, "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_CASU_reg/fc34fd97788acf1cc690744e002cad917889791f9ee1230f73e38aac10eb6750.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-1", "key": "fc9607d9f1f39574a745e679fdf2fab25f5f1a5e878cb4d719280af3443248c7", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "302305242faea44450e9d92717a05c328f99991b68afa7840b988ee92f1269fd", "generation_seconds_batch": 24.341178067028522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.964784622192383, "target_z": 0.16796877664261456, "ce": 6.09409761428833, "genuineness": 4.607763767242432, "identity": 0.5928120613098145, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/fc9607d9f1f39574a745e679fdf2fab25f5f1a5e878cb4d719280af3443248c7.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-0", "key": "fcbb99b75d1deee38b56559d586e3af4dbabc23580896f9c403553da4862737e", "bucket": "vn_S_CASU_reg", "phase": "wording", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "9fe2f1cc86f871b2ca2863cd6c36e5156d659da20113604d225b0489ec07d82b", "generation_seconds_batch": 16.456124094082043, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.673933982849121, "target_z": -0.8437500729934646, "ce": 6.125868320465088, "genuineness": 4.507347106933594, "identity": 0.8584567904472351, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/fcbb99b75d1deee38b56559d586e3af4dbabc23580896f9c403553da4862737e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "fceb68318e687581b953786a51f6609b4bb43964c88ee7ac880888ab8bd73743", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "2c7235e6c9950de909a964bef343da9b31405d589d590f2f3376fa81843e59e9", "generation_seconds_batch": 24.00280263996683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8613669872283936, "target_z": 0.0869140306566011, "ce": 6.300607204437256, "genuineness": 4.469690799713135, "identity": 0.7131958603858948, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/fceb68318e687581b953786a51f6609b4bb43964c88ee7ac880888ab8bd73743.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "fcefe2cfdd61ed61b0f8c9a23a5b5b2865e801b207fa7b35d9a9c46f2d0644b3", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "3bea1361fa057aafbc40418dc4edfba2f8e94e7bf3fbc027ef57b6f79d9803ee", "generation_seconds_batch": 22.563863244955428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2445995807647705, "target_z": -0.3964844291976474, "ce": 5.903473377227783, "genuineness": 3.9801597595214844, "identity": 0.20525288581848145, "hypothesis": "The door is still open. I'd put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/fcefe2cfdd61ed61b0f8c9a23a5b5b2865e801b207fa7b35d9a9c46f2d0644b3.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-es-0", "key": "fd074ab2b8ffae0327bfdd4a112915c10be895d4897d35dac1b817967489594c", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "c1f031c5da2db7106f91e6746c689eb44bc14e91da7af5c786a668eb02d1c399", "generation_seconds_batch": 21.745204703998752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.396610975265503, "target_z": -0.27734377226300666, "ce": 5.959072113037109, "genuineness": 3.998987913131714, "identity": 0.6153987050056458, "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_CASU_reg/fd074ab2b8ffae0327bfdd4a112915c10be895d4897d35dac1b817967489594c.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-1", "key": "fe6580d69ab4d53067ed3e9e072c13f785b16a10a8d85fd69b679008fe72a4e3", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I found the little box.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] It was beside the old window.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "8fdf8b0903122f8e9d8c063a7016fa9aadf8538c6557a9cb1d53131cb0d2a8cc", "generation_seconds_batch": 23.645970265963115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.154175758361816, "target_z": 0.3164062306567319, "ce": 6.23706579208374, "genuineness": 4.570107460021973, "identity": 0.5551100969314575, "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_CASU_reg/fe6580d69ab4d53067ed3e9e072c13f785b16a10a8d85fd69b679008fe72a4e3.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "fe70e39c18234e0925aafdc5090f4c8bb2745581e8a01066aaf9d5186e4f843e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "b1f8843c6f78a0ea21864d5ac8f53d0748e45da09cab94809acdf0eac244753f", "generation_seconds_batch": 24.327107973047532, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.197251796722412, "target_z": -0.43359379270117676, "ce": 6.054384231567383, "genuineness": 4.6203155517578125, "identity": 0.603253185749054, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/fe70e39c18234e0925aafdc5090f4c8bb2745581e8a01066aaf9d5186e4f843e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-nl-0", "key": "fecafd96d35dcbe58fb656d6954320399086849f7c511078ff5bbffc9f73454b", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "6fa49fd16b2ce04dc23d98a3b38fa1e3cccba72329aab6b857374e060980a3fe", "generation_seconds_batch": 18.261779900989495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8968777656555176, "target_z": 0.11474600656843077, "ce": 5.6532793045043945, "genuineness": 4.18099308013916, "identity": 0.4093068242073059, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/fecafd96d35dcbe58fb656d6954320399086849f7c511078ff5bbffc9f73454b.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": "ff0e93dc68f466b578891d2004d3bef8e4c927f2470cc6597b98f3c6e1edaeef", "bucket": "vn_S_CASU_reg", "phase": "heldout", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] I wrote the number on the envelope.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f6fab21df2c3dcab599d31f4e65a932481d4173df750e50beab2225c2e9190b", "generation_seconds_batch": 21.512020990019664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.111812114715576, "target_z": 0.28320324379686357, "ce": 6.348263263702393, "genuineness": 4.331617832183838, "identity": 0.40749475359916687, "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_CASU_reg/ff0e93dc68f466b578891d2004d3bef8e4c927f2470cc6597b98f3c6e1edaeef.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-fr-1", "key": "ff1a1c71c6e3f539759bd0a7149f209779f60bcabec61ec1d74f630b37f1432e", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "1259da04737186963f4dc9b286cfa9a9588448cd00b5660dc67df118575fc71f", "generation_seconds_batch": 23.40896548796445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6264970302581787, "target_z": -0.09716796569339867, "ce": 6.268836498260498, "genuineness": 4.770940780639648, "identity": 0.6748979687690735, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ff1a1c71c6e3f539759bd0a7149f209779f60bcabec61ec1d74f630b37f1432e.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-en-0", "key": "ff2b514ea0a3ace898e82f3fd206edee98b56dec79be6d211bbaeb86d76793bf", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] The door is still open.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] I put your coat on the chair.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "9e6f825e6cc0d52aef76c02eb3fd7d61171c233e1cdd97b28620de061ed80b6d", "generation_seconds_batch": 21.913127915002406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8999929428100586, "target_z": 0.11718756204444487, "ce": 6.0146708488464355, "genuineness": 4.381826400756836, "identity": 0.44902652502059937, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ff2b514ea0a3ace898e82f3fd206edee98b56dec79be6d211bbaeb86d76793bf.mp3" }, { "id": "vn_S_CASU_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_CASU_reg-de-0", "key": "ff9d9e3e560459e1f8c4c4ce4fdd3794fada74fcf523191a050b9b1598bad661", "bucket": "vn_S_CASU_reg", "phase": "grid", "candidate": "manual-e8c5567cec7f", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Emphasize the casual, unscripted nature of the speech; the speaker is rambling slightly, thinking out loud.\nSCRIPT:\n(Allow the pace to vary naturally, speeding up slightly on familiar phrases and slowing down for emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Incorporate subtle vocal texture shifts that suggest momentary distraction or shifting focus.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End the clip with a sense of lingering thought, as if the speaker is about to change the subject abruptly.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "5b780fc22e47311c13d777294b0dbad16bab9a82c0db7967a423073482822b0b", "generation_seconds_batch": 21.795407379977405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.000918388366699, "target_z": 0.1962890262769932, "ce": 5.689021587371826, "genuineness": 4.6203155517578125, "identity": 0.7207071781158447, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_CASU_reg/ff9d9e3e560459e1f8c4c4ce4fdd3794fada74fcf523191a050b9b1598bad661.mp3" }, { "key": "original-vn_S_CASU_reg-de-0-seed1701-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "354a1cb5df4cc6d984a1c5382643d0159b8c09991f51a8ee6f6e321d1397ebd8", "task_id": "vn_S_CASU_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.9031970500946045, "target_z": -0.6640625401464055, "ce": 6.062326908111572, "genuineness": 4.595211505889893, "identity": 0.7088621854782104, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/de/369d9e4726ca76c0b4bf9c45.mp3" }, { "key": "original-vn_S_CASU_reg-de-0-seed1701-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "d129349ffae10888c0a969350627e39d3f49ecb2062b03148318a3eea8a922d7", "task_id": "vn_S_CASU_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.095792531967163, "target_z": -1.2968750014598693, "ce": 6.109982967376709, "genuineness": 4.344170093536377, "identity": 0.6812130808830261, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/de/01f154b3968d83c40a6602b2.mp3" }, { "key": "original-vn_S_CASU_reg-de-0-seed1701-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "c5600248f9c83f13c6e9b520c46bad976fa322e5da221c5da8e7de172e7968fa", "task_id": "vn_S_CASU_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.8864654302597046, "target_z": -1.4609374948904574, "ce": 6.133810997009277, "genuineness": 4.482243061065674, "identity": 0.5378999710083008, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du wenn du bereit bist.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_CASU_reg/de/d4242536e8d94d2f26cfe940.mp3" }, { "key": "original-vn_S_CASU_reg-de-0-seed1701-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "31e34c41055a55b9fbe1ebdf53cdbcaffd626c46b9f9e0215e55f7a5dff7a333", "task_id": "vn_S_CASU_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.285183906555176, "target_z": -1.1484373605824827, "ce": 6.252951145172119, "genuineness": 4.369274139404297, "identity": 0.6387127637863159, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/de/9dda966ed310922206a87094.mp3" }, { "key": "original-vn_S_CASU_reg-de-0-seed1702-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "acff2fa78012d6e56ded9ec8ad321d33e4d5eef2b3e0684e2ed472c5726e2107", "task_id": "vn_S_CASU_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.6871063709259033, "target_z": -1.6171874218969928, "ce": 6.2767791748046875, "genuineness": 4.250029563903809, "identity": 0.6639021635055542, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/de/f76e55db6916837b8dd836bf.mp3" }, { "key": "original-vn_S_CASU_reg-de-0-seed1702-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "7c46e6ac0219e49d21ce6b379f50892829d8d2b2a3267f94ba893d0e758844e7", "task_id": "vn_S_CASU_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.035984754562378, "target_z": -1.3437500262776472, "ce": 6.23706579208374, "genuineness": 4.400654315948486, "identity": 0.5624878406524658, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/de/101fdb62f26f662ee846d558.mp3" }, { "key": "original-vn_S_CASU_reg-de-0-seed1702-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "292be94522ebf0ef3bd5237d88bb333b95806dde5b6c709c1ea2c0cec033a859", "task_id": "vn_S_CASU_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.3350234031677246, "target_z": -1.1093750890520266, "ce": 6.324435234069824, "genuineness": 4.645420074462891, "identity": 0.4816901683807373, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/de/2100de789d8ea3338c1e2a88.mp3" }, { "key": "original-vn_S_CASU_reg-de-0-seed1702-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "39372b480f8bbdb04b6c3dc95ed292ff2c27b14c3aabcb934b25ab2e3f8376d6", "task_id": "vn_S_CASU_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.1755361557006836, "target_z": -1.2343750306572552, "ce": 6.221180438995361, "genuineness": 4.394378662109375, "identity": 0.5142220258712769, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/de/516482b96ff4bc8cc0500e5d.mp3" }, { "key": "original-vn_S_CASU_reg-de-1-seed1701-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "82a12dd51e691dbe2142faf17b1cc9fc1822ad330bbb53d3eae809c7a3a0ab69", "task_id": "vn_S_CASU_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.3382277488708496, "target_z": -1.8906250043796078, "ce": 6.395919322967529, "genuineness": 4.143336772918701, "identity": 0.5797823071479797, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/de/aa0adc9aa9b3b4b824afaf22.mp3" }, { "key": "original-vn_S_CASU_reg-de-1-seed1701-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "95fe2ff262421ab90e6981ed3e2a1129ae3875c31f190cba4b240d3d8a3e8d29", "task_id": "vn_S_CASU_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.4977149963378906, "target_z": -1.7656250627743795, "ce": 6.268836498260498, "genuineness": 4.18099308013916, "identity": 0.21567508578300476, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/de/cd353455be54b8a3c42f5004.mp3" }, { "key": "original-vn_S_CASU_reg-de-1-seed1701-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "f128578cd0afb8669eb68b84fff74199d32b59765d9bcc5b96724d787d9d0ae8", "task_id": "vn_S_CASU_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.5276188850402832, "target_z": -1.7421875503654904, "ce": 6.395919322967529, "genuineness": 3.9927117824554443, "identity": 0.5348032712936401, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/de/7041d01cc23ad878e28c3f3e.mp3" }, { "key": "original-vn_S_CASU_reg-de-1-seed1701-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "3d3b80e38175b2fbeedbf6a6dd78fa94b100cd333e740ec66b9012d27f48e8c5", "task_id": "vn_S_CASU_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.9363051652908325, "target_z": -1.4218750364967323, "ce": 6.38797664642334, "genuineness": 4.256305694580078, "identity": 0.5115861892700195, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/de/0a197db41fba53091aedb262.mp3" }, { "key": "original-vn_S_CASU_reg-de-1-seed1702-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "6f69fa450278a97172a6139583c7cd78d5b570541d82adf5d9a3ad2346d6e71a", "task_id": "vn_S_CASU_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.6970741748809814, "target_z": -1.6093750423362094, "ce": 6.467403411865234, "genuineness": 4.193545341491699, "identity": 0.44871819019317627, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/de/083b6a477b1f4aeb61a1729e.mp3" }, { "key": "original-vn_S_CASU_reg-de-1-seed1702-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "3379bfaae7d96cd23191b4d271aea604296fe5aa37217d7d2a1d10ae05f13556", "task_id": "vn_S_CASU_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.258484125137329, "target_z": -1.953124975182222, "ce": 6.340320587158203, "genuineness": 4.074300289154053, "identity": 0.39625900983810425, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/de/c057c22ba8cabcad4e8ece90.mp3" }, { "key": "original-vn_S_CASU_reg-de-1-seed1702-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "d803b7bc46cc66556aebfd8b9e4a17044aff5d77a82e8b9a14609f3e8a95713f", "task_id": "vn_S_CASU_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.9263372421264648, "target_z": -1.4296875094891504, "ce": 6.41180419921875, "genuineness": 4.344170093536377, "identity": 0.323370099067688, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/de/35a090ae0322b819337f9574.mp3" }, { "key": "original-vn_S_CASU_reg-de-1-seed1702-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "6c8d1a1c1c2e5b00096e5a274e1227c5237f1b936d91f9729709fd3bdbb4bac4", "task_id": "vn_S_CASU_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.1556003093719482, "target_z": -1.2499999766420913, "ce": 6.427689552307129, "genuineness": 4.432034492492676, "identity": 0.5342328548431396, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/de/0fc27da847cee8befc69a01a.mp3" }, { "key": "original-vn_S_CASU_reg-en-0-seed1701-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "e8085f0c1c8c13374bd68f84184f7af53aec27a267f75956a9508264f38984a5", "task_id": "vn_S_CASU_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.6572024822235107, "target_z": -1.6406249343058819, "ce": 6.197352409362793, "genuineness": 3.9676074981689453, "identity": 0.583858847618103, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/en/e9d123130197e431cb9616c3.mp3" }, { "key": "original-vn_S_CASU_reg-en-0-seed1701-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "bebbd5c6b526d207e390f63233d49a373cc96d583705613d64f9bcd1405583d6", "task_id": "vn_S_CASU_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.9064013957977295, "target_z": -1.4453124554739867, "ce": 6.173524379730225, "genuineness": 3.936227321624756, "identity": 0.4209362864494324, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_CASU_reg/en/e3081e2f39aa09f673b1575e.mp3" }, { "key": "original-vn_S_CASU_reg-en-0-seed1701-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "90e77601f90519a447dfa28325dd922a1252450ff11ddbec50bc855d6db7fb65", "task_id": "vn_S_CASU_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.1755361557006836, "target_z": -1.2343750306572552, "ce": 6.260893821716309, "genuineness": 4.174716949462891, "identity": 0.3943116068840027, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/en/7a11529533df70fb040c0533.mp3" }, { "key": "original-vn_S_CASU_reg-en-0-seed1701-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "8dccfbeb144f3df3aa3ea9953b30f033a69e545447b527b510afd3ac1677445a", "task_id": "vn_S_CASU_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.4179713726043701, "target_z": -1.8281250335769936, "ce": 6.221180438995361, "genuineness": 3.8420867919921875, "identity": 0.48641741275787354, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/en/b2ab10c75b9cd24e62d73a18.mp3" }, { "key": "original-vn_S_CASU_reg-en-0-seed1702-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "3d04d9003deb9846f82db729f0a5b0a1fe780c0b1b8f6af1f4b1fe8b56adf46d", "task_id": "vn_S_CASU_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.7369461059570312, "target_z": -1.5781249635032677, "ce": 6.221180438995361, "genuineness": 4.218649387359619, "identity": 0.24530109763145447, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/en/b86124d1385fe87fe483418f.mp3" }, { "key": "original-vn_S_CASU_reg-en-0-seed1702-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "3d73569c8037f1ede7f72f94a32e574cc4cd5c4b6ffaa385e1c3ba994bba2ba4", "task_id": "vn_S_CASU_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.2784199714660645, "target_z": -1.9375000291973858, "ce": 6.173524379730225, "genuineness": 4.105680465698242, "identity": 0.37383347749710083, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_CASU_reg/en/348d7c1a16d48346ad060eb4.mp3" }, { "key": "original-vn_S_CASU_reg-en-0-seed1702-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "f00ecfa7177cdd66803dca876cd110af3672669ac3a5c1e0c50a9a686a169363", "task_id": "vn_S_CASU_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.2784199714660645, "target_z": -1.9375000291973858, "ce": 6.173524379730225, "genuineness": 3.8420867919921875, "identity": 0.32495182752609253, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_CASU_reg/en/abacc631dccaaaa2ca8f384c.mp3" }, { "key": "original-vn_S_CASU_reg-en-0-seed1702-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "166122931572f891f847c8e99802be20f9fd824b6f729119f85076df578e5a05", "task_id": "vn_S_CASU_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.1556003093719482, "target_z": -1.2499999766420913, "ce": 6.284721851348877, "genuineness": 4.105680465698242, "identity": 0.07405661046504974, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_CASU_reg/en/0976ebd0d87b051f60d0e28f.mp3" }, { "key": "original-vn_S_CASU_reg-en-1-seed1701-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "8070430dd0bebacdc8d1903f4d5164ecd47e32cfcf622adc9ea28963f748e6a9", "task_id": "vn_S_CASU_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.0028767585754395, "target_z": -0.5859374364956859, "ce": 6.221180438995361, "genuineness": 4.287685871124268, "identity": 0.4733208417892456, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/en/980ec43c6f5032b576fd8426.mp3" }, { "key": "original-vn_S_CASU_reg-en-1-seed1701-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "acc5a894346fc788732adbad9fa4dfbc1840ec44da1d0c8d28175ab27cc0eae1", "task_id": "vn_S_CASU_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 4.375962734222412, "target_z": 0.49023429196993407, "ce": 6.4515180587768555, "genuineness": 4.582659721374512, "identity": 0.2896835505962372, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/en/699d06b450762a4d116e6fb9.mp3" }, { "key": "original-vn_S_CASU_reg-en-1-seed1701-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "4f0d16681ba88fc4a7cde86372b473d6815c9dd754bd063bdca8ab82707897a8", "task_id": "vn_S_CASU_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.871335029602051, "target_z": 0.09472659708065383, "ce": 6.41180419921875, "genuineness": 4.6203155517578125, "identity": 0.29806196689605713, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/en/ae5e9e7e94bd14b907c92f5f.mp3" }, { "key": "original-vn_S_CASU_reg-en-1-seed1701-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "22ed1543f8eed0abe2cd81499200b35030d6d75822b40bb72a5ae05d3fd4d0aa", "task_id": "vn_S_CASU_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.9978926181793213, "target_z": -0.5898438131393469, "ce": 6.284721851348877, "genuineness": 4.3127899169921875, "identity": 0.3928492069244385, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/en/16d929a8ed24778668ab52ad.mp3" }, { "key": "original-vn_S_CASU_reg-en-1-seed1702-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "2907211bb67fa7dc524dc6a505285e876b9549f954c019c6a2de56ff208fdc62", "task_id": "vn_S_CASU_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.3268353939056396, "target_z": -0.3320311766415681, "ce": 6.197352409362793, "genuineness": 4.381826400756836, "identity": 0.3087626099586487, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/en/b9ef932c3c226124f6e4fd2a.mp3" }, { "key": "original-vn_S_CASU_reg-en-1-seed1702-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "a08fdc3644c15cef3bbf1800c4856d3bbbe6de0e11ff700eabf9c088786b878d", "task_id": "vn_S_CASU_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.600954055786133, "target_z": -0.11718756204444487, "ce": 6.292664527893066, "genuineness": 4.482243061065674, "identity": 0.19245094060897827, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/en/27dc2c1c68336e7c6a5ead8f.mp3" }, { "key": "original-vn_S_CASU_reg-en-1-seed1702-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "abd15a716452d3217b99133d0c539588125a4f859f08fc5b93f68abf8171ead2", "task_id": "vn_S_CASU_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 4.114304065704346, "target_z": 0.2851563386870594, "ce": 6.268836498260498, "genuineness": 4.300237655639648, "identity": 0.1822119802236557, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/en/801f1669883080c38699eca7.mp3" }, { "key": "original-vn_S_CASU_reg-en-1-seed1702-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "f13d2020c4d56a416bed57521c51a42f68c1a936d466d3e8d93bffb221df3259", "task_id": "vn_S_CASU_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.159872055053711, "target_z": -0.46289058978065334, "ce": 6.268836498260498, "genuineness": 4.444586753845215, "identity": 0.3869912922382355, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/en/3bdf46dab728f98c88ed8b09.mp3" }, { "key": "original-vn_S_CASU_reg-es-0-seed1701-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "08e5264860700c9942cfe0d9dde9250731f74e7d7ac9e829cad972b23801e534", "task_id": "vn_S_CASU_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.6671702861785889, "target_z": -1.6328125547450985, "ce": 5.927301406860352, "genuineness": 4.18099308013916, "identity": 0.6079401969909668, "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_CASU_reg/es/406b22ff32c04e7bc62b975d.mp3" }, { "key": "original-vn_S_CASU_reg-es-0-seed1701-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "b1e50ea423cfc250b23df06abd674183961aab4ce687f1dfbba4ed0f1a4e9b20", "task_id": "vn_S_CASU_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.9961129426956177, "target_z": -1.3750000116789542, "ce": 6.062326908111572, "genuineness": 4.18099308013916, "identity": 0.6142657995223999, "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_CASU_reg/es/b5b67d2061fc44bcb18ec8b0.mp3" }, { "key": "original-vn_S_CASU_reg-es-0-seed1701-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "bf6841fc0a8bb0fef74b90cebf8211c5a98c5abb76fe6287c2392a8d42d27316", "task_id": "vn_S_CASU_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.344991445541382, "target_z": -1.101562522627974, "ce": 6.054384231567383, "genuineness": 4.174716949462891, "identity": 0.5668513178825378, "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_CASU_reg/es/8cbedaf3875df13144be26e4.mp3" }, { "key": "original-vn_S_CASU_reg-es-0-seed1701-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "2566c52b47f8544e8987acf9077bc3c9581cfa9518c1e10ccc86d2445d448039", "task_id": "vn_S_CASU_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.255280017852783, "target_z": -1.1718748729913717, "ce": 5.9431867599487305, "genuineness": 4.017816066741943, "identity": 0.3879604637622833, "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_CASU_reg/es/e0b4a65576f145df29c01213.mp3" }, { "key": "original-vn_S_CASU_reg-es-0-seed1702-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "365584b161c8d06842b2e4db71fb4d3cad577ea6dafce7f8ff5ac5d77764fec9", "task_id": "vn_S_CASU_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.876497507095337, "target_z": -1.4687499678828757, "ce": 5.895530700683594, "genuineness": 4.005263805389404, "identity": 0.4572364091873169, "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/audio/vn_S_CASU_reg/es/82e5c19c26a990c2b1b31f1f.mp3" }, { "key": "original-vn_S_CASU_reg-es-0-seed1702-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "de9d6a3623a966a8fa0f5e25176b206e68ba1c33651c268f793e2049d890c9cb", "task_id": "vn_S_CASU_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.1755361557006836, "target_z": -1.2343750306572552, "ce": 5.808161735534668, "genuineness": 4.325342178344727, "identity": 0.5236048698425293, "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_CASU_reg/es/8eb6b2f91a727c2a82a31d3f.mp3" }, { "key": "original-vn_S_CASU_reg-es-0-seed1702-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "6beb0ebb36cb79e39656f13ad2e1bcbddbeca5f2514651c0b9c2cf7bee90faa5", "task_id": "vn_S_CASU_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.444671154022217, "target_z": -1.0234374189772544, "ce": 6.062326908111572, "genuineness": 4.331617832183838, "identity": 0.6571692228317261, "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_CASU_reg/es/77c1f0a2d281b6a7c25b5358.mp3" }, { "key": "original-vn_S_CASU_reg-es-0-seed1702-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "456d1e2b7f200a629dbbcdddb84f19a288b7d91a52ba4bd5655c9ada8ae17918", "task_id": "vn_S_CASU_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.145632266998291, "target_z": -1.257812543066144, "ce": 6.173524379730225, "genuineness": 3.9425034523010254, "identity": 0.46516120433807373, "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_CASU_reg/es/0a5dddf4ab3aeb3a465d6fca.mp3" }, { "key": "original-vn_S_CASU_reg-es-1-seed1701-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "91f0bebfb64176d444af92e3e5439eee3cbfc724bf98a400c86fee2d817aaea3", "task_id": "vn_S_CASU_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.296931505203247, "target_z": -0.3554686890504571, "ce": 6.173524379730225, "genuineness": 4.770940780639648, "identity": 0.7910032868385315, "hypothesis": "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/audio/vn_S_CASU_reg/es/4241d4ff4f0a3b856ffc3c10.mp3" }, { "key": "original-vn_S_CASU_reg-es-1-seed1701-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "400b649a0b2ceef30be7f3c8a37ee2e0aec7eb807ddfa23cbfff72324b60ff34", "task_id": "vn_S_CASU_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.833421468734741, "target_z": -0.718749944524967, "ce": 6.006728172302246, "genuineness": 4.419482707977295, "identity": 0.42981410026550293, "hypothesis": "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/audio/vn_S_CASU_reg/es/04cff0bd532e3bbc5e1767ed.mp3" }, { "key": "original-vn_S_CASU_reg-es-1-seed1701-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "47b730b22cba8dedd4db2a0e379c4e324d4a81e569fc73f37ae4ddea42cbfac6", "task_id": "vn_S_CASU_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.1175081729888916, "target_z": -0.4960937635037909, "ce": 6.133810997009277, "genuineness": 4.607763767242432, "identity": 0.6011806726455688, "hypothesis": "He encontrado, la cajita estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/es/0f903b73191359350da862b9.mp3" }, { "key": "original-vn_S_CASU_reg-es-1-seed1701-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "9eaa60b6af57fba7edd8295de0ca322392a911b80ea32eefbd4dc5691a4212af", "task_id": "vn_S_CASU_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.1349520683288574, "target_z": -0.48242191240915056, "ce": 5.982900142669678, "genuineness": 4.6705241203308105, "identity": 0.6676203012466431, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/es/d08a15be13420f9e00682a5e.mp3" }, { "key": "original-vn_S_CASU_reg-es-1-seed1702-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "5e15f83d88da9402282fdd0b61092da438de4318d8e68c64cb7e1f261d0392aa", "task_id": "vn_S_CASU_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.2652478218078613, "target_z": -1.164062493430588, "ce": 6.0305562019348145, "genuineness": 4.306513786315918, "identity": 0.6385737657546997, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/es/ce11442d8b2fd373b3994c9c.mp3" }, { "key": "original-vn_S_CASU_reg-es-1-seed1702-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "e6d9b7028da85aa9f5da2075e8bc3b030ea2619d373523e234a83a7864b7f8ab", "task_id": "vn_S_CASU_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.6839020252227783, "target_z": -0.8359375065694118, "ce": 6.133810997009277, "genuineness": 4.7081804275512695, "identity": 0.6344060897827148, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/es/a6a3f6a1fa4fd6a801abdfd7.mp3" }, { "key": "original-vn_S_CASU_reg-es-1-seed1702-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "103e4c7612543347ecf06616dd90d959e79d9359bfaee6d38645919a2c3100eb", "task_id": "vn_S_CASU_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.504478931427002, "target_z": -0.9765623941594764, "ce": 6.165581703186035, "genuineness": 4.582659721374512, "identity": 0.611096978187561, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/es/31f2793692bb7830caf8e726.mp3" }, { "key": "original-vn_S_CASU_reg-es-1-seed1702-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "b31da9e2ace412a28b7bcb570694706fb5111749198af08e83d2491681db7338", "task_id": "vn_S_CASU_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.205440044403076, "target_z": -1.2109375182483662, "ce": 5.998785495758057, "genuineness": 4.432034492492676, "identity": 0.6178035140037537, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/es/cd592ffc6ee11d3158ec48c5.mp3" }, { "key": "original-vn_S_CASU_reg-fr-0-seed1701-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "b142b0353d8a7ea5e4c518886964fef64542242fb294e5c34f883f0ae4fd49e8", "task_id": "vn_S_CASU_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.6970741748809814, "target_z": -1.6093750423362094, "ce": 6.086154937744141, "genuineness": 4.086852550506592, "identity": 0.7047086954116821, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/fr/e4a9b866045fff7093e1cf63.mp3" }, { "key": "original-vn_S_CASU_reg-fr-0-seed1701-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "ef089ad18d9474a03166c7e2deb624d00c967842ff5eee02fe873d0c7292aedf", "task_id": "vn_S_CASU_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.7935495376586914, "target_z": -0.7500000233579086, "ce": 6.284721851348877, "genuineness": 4.469690799713135, "identity": 0.5987257957458496, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/fr/dfe5759484d5b53051ee27d5.mp3" }, { "key": "original-vn_S_CASU_reg-fr-0-seed1701-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "af90e019aeb5917456498ca377f489cdee3cb4b95e9feb564a9f834385faeb66", "task_id": "vn_S_CASU_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.4745750427246094, "target_z": -0.9999999065683653, "ce": 6.0146708488464355, "genuineness": 4.645420074462891, "identity": 0.5280970931053162, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/fr/7ae5469f01e177f78eafeb1b.mp3" }, { "key": "original-vn_S_CASU_reg-fr-0-seed1701-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "3709383801e8350d5f45d968db9228a334fc83ff8717fefb979fe33ba960ff1f", "task_id": "vn_S_CASU_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.5892064571380615, "target_z": -0.9101562335764705, "ce": 6.173524379730225, "genuineness": 4.432034492492676, "identity": 0.6255308389663696, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/fr/6a3112c390a01d7db1e5d053.mp3" }, { "key": "original-vn_S_CASU_reg-fr-0-seed1702-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "0642a0a351338c11bb9f2ba6bbb3ca9204b0890922cbff7a006edf3cccafc9d4", "task_id": "vn_S_CASU_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.2154078483581543, "target_z": -1.2031251386875828, "ce": 6.173524379730225, "genuineness": 4.2374773025512695, "identity": 0.6262437105178833, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/fr/fa7f170b983366d096568c69.mp3" }, { "key": "original-vn_S_CASU_reg-fr-0-seed1702-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "d8e90a83e40355a09abee1830ca3bef30a6f352544d3e5f5bab58b5904810ef3", "task_id": "vn_S_CASU_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.9363051652908325, "target_z": -1.4218750364967323, "ce": 5.982900142669678, "genuineness": 4.369274139404297, "identity": 0.6794968843460083, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/fr/f8269e2bb2f3f39ed283f49c.mp3" }, { "key": "original-vn_S_CASU_reg-fr-0-seed1702-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "51acd9f370a1cd8452b838b9006d1e06957d05170d5272f99eacd4580ed0a0ac", "task_id": "vn_S_CASU_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.6888861656188965, "target_z": -0.8320311299257508, "ce": 6.086154937744141, "genuineness": 4.457139015197754, "identity": 0.6364963054656982, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/fr/9cbb4fd90ba9f97cb170744b.mp3" }, { "key": "original-vn_S_CASU_reg-fr-0-seed1702-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "aeccb87701a2169b26193cbfc911ea463f44c2af000d71415ba890a0175f7126", "task_id": "vn_S_CASU_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 1.9363051652908325, "target_z": -1.4218750364967323, "ce": 6.078212261199951, "genuineness": 3.9927117824554443, "identity": 0.68418288230896, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/fr/7620e7f5d718b51b4c4ba5f9.mp3" }, { "key": "original-vn_S_CASU_reg-fr-1-seed1701-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "e4ae92fa97ef11b4af83614b49762c84ff13d44ad309bcad0ba3fb3910325811", "task_id": "vn_S_CASU_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.137444019317627, "target_z": -0.4804688175189547, "ce": 6.332377910614014, "genuineness": 4.720732688903809, "identity": 0.5550294518470764, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/fr/31010c3b4031e57d06abacb9.mp3" }, { "key": "original-vn_S_CASU_reg-fr-1-seed1701-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "b80b2979ee409318f1afee2e70b624855778a6ac84410d39eff6fe1858f2ee1e", "task_id": "vn_S_CASU_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.4247350692749023, "target_z": -1.0390625518253598, "ce": 6.213237762451172, "genuineness": 4.6203155517578125, "identity": 0.7162526845932007, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/fr/cb367f5063914cf4e4998c97.mp3" }, { "key": "original-vn_S_CASU_reg-fr-1-seed1701-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "7961cba832782e01f52da236958bc78c86742bc751f32647076d05af764657e7", "task_id": "vn_S_CASU_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.813485622406006, "target_z": -0.7343748905098032, "ce": 6.23706579208374, "genuineness": 4.545003414154053, "identity": 0.7814626693725586, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/fr/dd4f4b4d43e817b8ff6ff4ef.mp3" }, { "key": "original-vn_S_CASU_reg-fr-1-seed1701-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "f58aec573bfa5fdb3124ce05eba693c388243ebfd090cac7c720d19d04c08d14", "task_id": "vn_S_CASU_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.195472240447998, "target_z": -1.2187498978091496, "ce": 6.149696350097656, "genuineness": 4.306513786315918, "identity": 0.6330538988113403, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/fr/7c9215fadc9c1f086126340a.mp3" }, { "key": "original-vn_S_CASU_reg-fr-1-seed1702-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "0d3271ab85d7420b2cf2790137b3acc4dc294c605f31b157d8f71525395b6412", "task_id": "vn_S_CASU_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.2620434761047363, "target_z": -0.38281257810300706, "ce": 6.260893821716309, "genuineness": 4.833701133728027, "identity": 0.8214512467384338, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/fr/ce430c4f378cb869113c7e4d.mp3" }, { "key": "original-vn_S_CASU_reg-fr-1-seed1702-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "cecdcedd0cbceceb9f7f6f4aee368c488037472a3747f6032a7453c8b0a74f2d", "task_id": "vn_S_CASU_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.3268353939056396, "target_z": -0.3320311766415681, "ce": 6.252951145172119, "genuineness": 4.7081804275512695, "identity": 0.5443875193595886, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/fr/a293a9d53292901863d36a6f.mp3" }, { "key": "original-vn_S_CASU_reg-fr-1-seed1702-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "992c2f3d4a8798135592b8f058b1b40aa846481b9945ea9e9bca28de03380f55", "task_id": "vn_S_CASU_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.7598185539245605, "target_z": 0.007324292701503773, "ce": 6.173524379730225, "genuineness": 4.595211505889893, "identity": 0.5916279554367065, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/fr/7022ba34193349453325b901.mp3" }, { "key": "original-vn_S_CASU_reg-fr-1-seed1702-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "a6fc9271531b65b029e0e7497de8b05ef5c3420719657de1632245a120a23a60", "task_id": "vn_S_CASU_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.763645648956299, "target_z": -0.7734375357667976, "ce": 5.855817794799805, "genuineness": 4.419482707977295, "identity": 0.6935372352600098, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/fr/edaddf1743d60a55a67672a8.mp3" }, { "key": "original-vn_S_CASU_reg-nl-0-seed1701-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "44e053aa53872c8238a31e4e3f208baa67d2c69821e3c7574af78e0af74c446b", "task_id": "vn_S_CASU_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.3118834495544434, "target_z": -0.34374993284601263, "ce": 6.141753673553467, "genuineness": 4.362998008728027, "identity": 0.7146576642990112, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/nl/23843f3c82061e5828887de1.mp3" }, { "key": "original-vn_S_CASU_reg-nl-0-seed1701-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "62464d9084b54601319a8549e75678eadd71b041bef7fc051d8791a5c1433bfb", "task_id": "vn_S_CASU_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.1075401306152344, "target_z": -0.5039063299278437, "ce": 5.990842819213867, "genuineness": 4.18099308013916, "identity": 0.6253570318222046, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/nl/ecd2fab83106df9b3502c9de.mp3" }, { "key": "original-vn_S_CASU_reg-nl-0-seed1701-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "bb5e450a3b7bcc19661e32d212d66f0b2e7f548165205236e6bdb86d335900cc", "task_id": "vn_S_CASU_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.4925525188446045, "target_z": -0.2021484978108501, "ce": 6.1020402908325195, "genuineness": 4.300237655639648, "identity": 0.333489328622818, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/nl/772a7a6377aa8c972abfad99.mp3" }, { "key": "original-vn_S_CASU_reg-nl-0-seed1701-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "2cd67e17e8bb082f20ea7409dc17f98ceeeb7c872ffab11d88c716110bb55fd7", "task_id": "vn_S_CASU_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9562411308288574, "target_z": -1.4062499970802613, "ce": 6.078212261199951, "genuineness": 4.18099308013916, "identity": 0.5978378057479858, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/nl/af3d9ec481f74c009ac4fc12.mp3" }, { "key": "original-vn_S_CASU_reg-nl-0-seed1702-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "5a685b61a547c80731cddfeb49f950e87b6ef1b9d5c4708692d211d42b75f474", "task_id": "vn_S_CASU_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.0078606605529785, "target_z": -0.582031246715294, "ce": 6.117925643920898, "genuineness": 4.519899368286133, "identity": 0.7782447338104248, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/nl/f2d8c84950e334e77c75268e.mp3" }, { "key": "original-vn_S_CASU_reg-nl-0-seed1702-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "3e5fe1a5dd17fbe07e15519b57e5c866a301f7b2cf8bf9d1b805a9947832a966", "task_id": "vn_S_CASU_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5375869274139404, "target_z": -1.7343749839414377, "ce": 5.998785495758057, "genuineness": 4.1998209953308105, "identity": 0.6244016885757446, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/nl/dab994c2b89775ddcb304ab3.mp3" }, { "key": "original-vn_S_CASU_reg-nl-0-seed1702-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "aed73cf8660114ed9f745913995e6309620d69d19c2ff8baee828878f6271c08", "task_id": "vn_S_CASU_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.557344436645508, "target_z": -0.15136709634941115, "ce": 6.0146708488464355, "genuineness": 4.381826400756836, "identity": 0.6306262016296387, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/nl/5848721a4f0b91fbe036e03a.mp3" }, { "key": "original-vn_S_CASU_reg-nl-0-seed1702-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "093e42bddb023c7522eeb75cd42bbf4d10ba936cca837a9f5333f723c45e1e9f", "task_id": "vn_S_CASU_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.356739044189453, "target_z": -0.3085938510959484, "ce": 6.157639026641846, "genuineness": 4.319066047668457, "identity": 0.6613445281982422, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_CASU_reg/nl/1fd3b5046ebd9d68d467d1ae.mp3" }, { "key": "original-vn_S_CASU_reg-nl-1-seed1701-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "3cfe433cd66ae43b44a6e3e8e5702348f6e3684c6f9ab573bdf8e8107ec09f86", "task_id": "vn_S_CASU_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9961129426956177, "target_z": -1.3750000116789542, "ce": 6.181467056274414, "genuineness": 4.300237655639648, "identity": 0.6923670768737793, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/nl/8a922519f85e1f68cab4874b.mp3" }, { "key": "original-vn_S_CASU_reg-nl-1-seed1701-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "08a796f06eb4afcae02ab7452b756a9c6f4c1e6907cfc3949f70edde0e5f9473", "task_id": "vn_S_CASU_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 4.161651611328125, "target_z": 0.3222655153273195, "ce": 5.967014789581299, "genuineness": 4.394378662109375, "identity": 0.39741095900535583, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/nl/73f56500eb9fc8a9763ddbaf.mp3" }, { "key": "original-vn_S_CASU_reg-nl-1-seed1701-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "3e6d88fc4e81f002effdc0a7b1d45deb9f93605146f9034e0ee6489445c73ae0", "task_id": "vn_S_CASU_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.478846549987793, "target_z": -0.21289070657019663, "ce": 6.09409761428833, "genuineness": 4.325342178344727, "identity": 0.6820990443229675, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/nl/3eed6ebc7b14ca357ba38a9b.mp3" }, { "key": "original-vn_S_CASU_reg-nl-1-seed1701-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "edf6041d82084b3dee2cff5eb132994cb9558378eb038d7e78aac5118d21b215", "task_id": "vn_S_CASU_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9594454765319824, "target_z": -2.1874999124078425, "ce": 6.0146708488464355, "genuineness": 3.8546390533447266, "identity": 0.6754474639892578, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/nl/bbb9498d92a40d83d7e12448.mp3" }, { "key": "original-vn_S_CASU_reg-nl-1-seed1702-scale0.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "4c3181ba1a4501b2042fff871f844e535fdc2733d9c43c62c539061996806e22", "task_id": "vn_S_CASU_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.037764549255371, "target_z": -0.5585937343064051, "ce": 6.24500846862793, "genuineness": 4.507347106933594, "identity": 0.5393973588943481, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/nl/1fcfcc74112de9d7a05a7166.mp3" }, { "key": "original-vn_S_CASU_reg-nl-1-seed1702-scale0.5", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "4980cdb26c5c9bf3feed1e6e702d88fc49d2d8a0dd47665b91eb597c9b53516e", "task_id": "vn_S_CASU_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.5685582160949707, "target_z": -0.1425781693435297, "ce": 6.006728172302246, "genuineness": 4.419482707977295, "identity": 0.7080590724945068, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/nl/79c78018f1a4a37bcbdc9405.mp3" }, { "key": "original-vn_S_CASU_reg-nl-1-seed1702-scale0.75", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "d6cdc501b63e27852693cda9dbb5954ca59e0ec7f1123e4742c05fd3f84192ab", "task_id": "vn_S_CASU_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 2.444671154022217, "target_z": -1.0234374189772544, "ce": 6.062326908111572, "genuineness": 4.325342178344727, "identity": 0.7487696409225464, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_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_CASU_reg/nl/20d308cd0b665cda87a33a66.mp3" }, { "key": "original-vn_S_CASU_reg-nl-1-seed1702-scale1.0", "bucket": "vn_S_CASU_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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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 Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Casual 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": "21d180b53d8a097054f4727e41f9749a1139246ddfda9059c3105e511968ae9b", "mp3_sha256": "320d6eef39634656efec4605552dc605838d62221d8016164c55e948f5b19ae3", "task_id": "vn_S_CASU_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Casual Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Casual 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 Casual 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 Casual 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": 3.997180461883545, "target_z": 0.1933593839416994, "ce": 5.998785495758057, "genuineness": 4.268857479095459, "identity": 0.7887685298919678, "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_CASU_reg/nl/edd2a2d334c5fa046e07a353.mp3" } ], "source_standard_audio_unchanged": true }